Forecasting Techniques 95

Notes

Unit 3: Forecasting Techniques

Structure:
3.1 Introduction
3.2 Sample Correlation Analysis
3.3 Types of Correlation
3.4 Methods of Determining Correlation
3.5 Scatter Diagram
3.4 Methods of Determining Correlation
3.6 Karl Pearsons Co-efficient of Correlation
3.7 Rank Correlation
3.8 Regression Analysis
3.9 Time Series Analysis
3.10 Trend Analysis
3.11 Components of Time Series
3.12 Methods of Finding Trend
3.13 Moving Average Methods
3.14 Least Squares Methods
3.15 Summary
3.16 Check Your Progress
3.17 Questions and Exercises
3.18 Key Terms
3.19 Check Your Progress: Answers
3.20 Case Study
3.21 Further Readings
3.22 Bibliography

Objectives

After studying this unit, you should be able to understand:

 Sample Correlation Analysis
 Regression Analysis
 Overview of Time Series Analysis
 Concept of Trend Analysis
 Cyclical Analysis
 Concept of Seasonal Analysis, Irregular Variation

Amity Directorate of Distance and Online Education

96 Quantitative Techniques in Management

Notes 3.1 Introduction

Francis Galton was the first person to measure correlation, originally termed “co-
relation,” which actually makes sense considering you’re studying the relationship between
a couple of different variables. In Co-Relations and Their Measurement, he said “The
statures of kinsmen are co-related variables; thus, the stature of the father is correlated
to that of the adult son,..and so on; but the index of co-relation … is different in the different
cases.” It’s worth noting though that Galton mentioned in his paper that he had borrowed
the term from biology, where “Co-relation and correlation of structure” was being used but
until the time of his paper it hadn’t been properly defined.
In 1892, British statistician Francis Ysidro Edgeworth published a paper called
“Correlated Averages,” Philosophical Magazine, 5th Series, 34, 190-204 where he used
the term “Co-efficient of Correlation.” It wasn’t until 1896 that British mathematician Karl
Pearson used “Coefficient of Correlation” in two papers: Contributions to the Mathematical
Theory of Evolution and Mathematical Contributions to the Theory of Evolution. III.
Regression, Heredity and Panmixia
Correlation is a bivariate analysis that measures the strengths of association between
two variables. In statistics, the value of the correlation co-efficient varies between +1 and
-1. When the value of the correlation coefficient lies around ± 1, then it is said to be
a perfect degree of association between the two variables. As the correlation coefficient
value goes towards 0, the relationship between the two variables will be weaker. The goal
of a correlation analysis is to see whether two measurement variables co vary and to
quantify the strength of the relationship between the variables. Correlational analysis and
research is useful in providing links between variables that can further be investigated,
however as correlation does not infer cause, this type of research can also be affected
by mediating factors, making it lack internal validity.

3.2 Sample Correlation Analysis

Correlation is a statistical measure that indicates the extent to which two or more
variables fluctuate together. A positive correlation indicates the extent to which those
variables increase or decrease in parallel; a negative correlation indicates the extent to
which one variable increases as the other decreases.
Meaning of Correlation
Correlation means that between two series or groups of data there exists some causal
connections.” At another place, he says, “If it is proved true that in a large number of
instances two variables tend always to fluctuate in the same or in opposite directions,
we consider that the fact is established and that a relationship exists. The relationship
is called correlation.
Definitions of Correlation
According to E. Davenport, “the whole subject of correlation refers to that
interrelation between separate character by which they tend, in some degree atleast, to
move together.”
According to Prof. Boddington, “whenever some definite connection exists between
the two or more groups, classes or series of data, there is said to be correlation.”
According to Croxton and Cowden,“When the relationship is of quantitative nature,
the appropriate statistical tool for discovering and measuring the relationship and
expressing it in a brief formula is known as correlation.”

Amity Directorate of Distance and Online Education

Forecasting Techniques 97

Tippett states that, “the effect of correlation is to reduce the range of uncertainty Notes
of our prediction.”
In the words L. R. Conner “if two or more quantities vary in sympathy so that
movements in one tend to be accompanied by corresponding movement in the other then
they are said to be correlated.”
The statistical technique which deals with the association between two or more
variables is known as correlation analysis.
A. M. Tuttle defines correlation analysis as the Co-variation between two or more
variables.

Uses of Correlation

There are three main uses for correlation and regression:

1. It is used to test hypotheses about cause-and-effect relationships. In this case,
the experimenter determines the values of the X-variable and sees whether
variation in X causes variation in Y. For example, giving people different amounts
of a drug and measuring their blood pressure.
2. The second main use for correlation and regression is to see whether two
variables are associated, without necessarily inferring a cause-and-effect
relationship. In this case, neither variable is determined by the experimenter;
both are naturally variable. If an association is found, the inference is that
variation in X may cause variation in Y or variation in Y may cause variation
in X or variation in some other factor may affect both X and Y.
3. The third common use of linear regression is estimating the value of one variable
corresponding to a particular value of the other variable.

Advantages of Correlation
1. Correlation research allows researchers to collect much more data than
experiments.
2. Correlation research is that it opens up a great deal of further research to other
scholars.
3. It allows researchers to determine the strength and direction of a relationship
so that later studies can narrow the findings down and if possible, determine
causation experimentally.
4. Gain quantitative data which can be easily analysed.
5. No manipulation of behaviour is required.
6. The correlation coefficient can readily quantify observational data.

Disadvantages of Correlation
1. Correlation research only uncovers a relationship; it cannot provide a conclusive
reason for why there's a relationship.
2. A correlative finding doesn't reveal which variable influences the other. For
example, finding that wealth correlates highly with education doesn't explain
whether having wealth leads to more education or whether education leads to
more wealth.
3. Reasons for either can be assumed, but until more research is done, causation
can't be determined.

Amity Directorate of Distance and Online Education

98 Quantitative Techniques in Management

Notes 4. Here, a third, unknown variable might be causing both. For instance, living in
the City of Bangalore can lead to both wealth and education.

3.3 Types of Correlation

Types of Correlation

On the basis of On the basis of On the basis of

Direction Number of Sets Change

Same Opp. Partial Only Two More than Linear Non-Linear

(Positive (Negative (One Sets Two sets (Straight (Curve Type)
or Direct) or Indirect) depending (Simple) (Multiple) Line)
on other)

I. On the basis of Direction

1. Positive Correlation
The correlation is said to be positive correlation if the values of two variables changing
with same direction.
Example: Production expenses and sales, Height and weight, water consumption
and temperature, study time and grades etc.
2. Negative Correlation
The correlation is said to be negative correlation when the values of variables change
with opposite direction. Example: Price and Quantity demanded, alcohol consumption and
driving ability etc.
Direction of the Correlation
Positive relationship – Variables change in the same direction.
As X is increasing - Y is increasing.
As X is decreasing - Y is decreasing.
Example, As height increases, so does weight.
Negative relationship – Variables change in opposite directions.
As X is increasing - Y is decreasing.
As X is decreasing - Y is increasing.
Example, As TV time increases, grades decrease.
3. Partial Correlation
In partial correlation more than two variables are recognised but only two variables
influence each other, the effect of other influencing variable is kept constant. In the above
example, if we limit our correlation analysis of yield and rainfall keeping fertilizer variable
as constant to becomes a problem of partial correlation.

Amity Directorate of Distance and Online Education

Forecasting Techniques 99

II. On the basis of Number of Sets Notes

1. Simple Correlation
When only two variables are studied it is a case of simple correlation.
2. Multiple Correlation
When more than three variables one Studied it is known as multiple correlation. For
examples, When we study the relationship between the yield of rice per acre and both
the amount of rainfall and the amount of fertilizer used it is a case of multiple correlation.

III. On the basis of Change

1. Linear Correlation
Correlation is said to be linear when the amount of change in one variable tends
to bear a constant ratio to the amount of change in the other. The graph of the variables
having a linear relationship will form a straight line.
Example:
X = 1, 2, 3, 4, 5, 6, 7, 8
Y = 5, 7, 9, 11, 13, 15, 17, 19
Y = 3 + 2x
2. Non Linear correlation
The correlation would be non linear if the amount of change in one variable does
not bear a constant ratio to the amount of change in the other variable
X 1 2 3 4 5
Y 7 14 21 28 35

3.4 Methods of Determining Correlation

The various methods of studying correlation are as follows:
1. Scatter diagram
2. Karl Pearsons Coefficient of correlation
3. Rank Correlation

3.5 Scatter Diagram

Scatter Diagram is a graph of observed plotted points where each points represents
the values of X & Y as a co-ordinate. It portrays the relationship between these two variables
graphically.
Sl. No. Maths Statistics
1. 55 60
2. 70 65
3. 35 50
4. 40 60
5. 65 75

Amity Directorate of Distance and Online Education

100 Quantitative Techniques in Management

Notes 6. 40 70 y
Scatter
70 Diagram
7. 60 50
60
8. 20 40
50

Statistics
9. 30 60 Estimating Line
40
10. 50 30
30
11. 10 30 Low Degree of
20
Positive
12. 20 10 Correlation
10
0 x

10

20

30

40

50

60

70
Advantages of Scatter Diagram Maths

1. It is a very simple and non mathematical method.

2. It is not influenced by the size of extreme item.
3. First step in investing the relationship between two variables.

Disadvantage of Scatter Diagram

1. It cannot be adopted the exact degree of correlation.

3.6 Karl Pearsons Co-efficient of Correlation

Karl Pearson’s Coefficient of Correlation denoted by- ‘r’ (-1 = r = +1). The coefficient
of correlation ‘r’ measure the degree of linear relationship between two variables say x
& y.
Degree of Correlation is expressed by a value of Coefficient. Direction of change is
Indicated by sign (- ve) or (+ ve).

 xy
When deviation taken from actual mean: r =
 x 2   y2

( x )( y)
xy 
n
When deviation taken from an assumed mean: r = ( x ) 2 ( y ) 2
x 2  y 2 
n n

Interpretation of Correlation Co-efficient (r)

The value of correlation coefficient ‘r’ ranges from -1 to +1

a) If r = +1, then the correlation between the two variables is said to be perfect
and positive.
b) If r = -1, then the correlation between the two variables is said to be perfect
and negative.
c) If r = 0, then there exists no correlation between the variables.

Amity Directorate of Distance and Online Education

Forecasting Techniques 101

Properties of Correlation Co-efficient Notes

a) The correlation co-efficient lies between -1 and +1 symbolically (- 1  r 1)
b) The correlation co-efficient is independent of the change of origin and scale.
c) The co-efficient of correlation is the geometric mean of two regression co-
efficient. r = bxy * byx
d) The one regression coefficient is (+ve) other regression coefficient is also (+ve)
correlation coefficient is (+ve).

Co-efficient of Determination

The convenient way of interpreting the value of correlation coefficient is to use of

square of coefficient of correlation which is called Coefficient of Determination.
The Co-efficient of Determination = r 2 .
Suppose: r = 0.9,

r 2 = 0.81 this would mean that 81% of the variation in the dependent variable has
been explained by the independent variable.
The maximum value of r 2 is 1 because it is possible to explain all of the variation
in y but it is not possible to explain more than all of it.
Coefficient of Determination = Explained variation / Total variation.

Merits of Karl Pearson’s Co-efficient of Correlation

1. This method indicates the presence or absence of correlation between two
variables and gives the exact degree of their correlation.
2. To ascertain the direction of the correlation positive or negative.
3. This method has many algebraic properties for which the calculation of co-
efficient of correlation and other related factors are made easy.

Demerits of Karl Pearson’s Co-efficient of Correlation

1. It is more difficult to calculate than other methods of calculations.

2. It is much affected by the values of the extreme items.
3. It is very much likely to be misinterpreted in case of homogeneous data.

Procedure for Computing the Correlation Co-efficient

1. Calculate the mean of the two series ‘x’ & ‘y’.

2. Calculate the deviations ‘x’ & ‘y’ in two series from their respective mean.
3. Square each deviation of ‘x’ & ‘y’ then obtain the sum of the squared deviation
i.e.,  x 2 &  y 2
4. Multiply each deviation under x with each deviation under y & obtain the product
of ‘xy’.
5. Then obtain the sum of the product of x, y i.e.,  xy
6. Substitute the value in the formula.

Amity Directorate of Distance and Online Education

102 Quantitative Techniques in Management

Notes Probable Error

Probable error of Co-efficient of Correlation gives us the two limits within which the
co-efficient of correlation of Series selected at random from the same universe is likely
to fall. The formula for the probable error of r is as follows.

1  r2
0.6745
N

Where, r = Coefficient of Correlation

N = Number of pairs of observations
Significance of Probable Error
Probable error is useful in the following interpretation:
1) If the value of ‘N’ is less than the probable error there is no evidence of correlation
i.e., The value of ‘r’ is not at all significant.
2) If the value of ‘r’ is more than 6 times the probable error, the Co-efficient of
Correlation is practically certain, i.e., The value of ‘r’ is Significant.
3) By adding and Subtracting the value of probable error from the coefficient of
correlation we get respectively, the upper and lower limits within Which co-
efficient of correlation in the population can be expected to be. symbolically.
P = r ± P.E
Where  (rho) devotes Correlation in the population
The Conditions necessary for the use of probable error
The measure of probable error can be Properly used only When the following three
conditions exist:
1) The data must approximate a normal frequency curve, i.e. bell shaped curve.
2) The Statistical measure for which the P.E is computed must have been
calculated from a Sample.
3) The sample must have been Selected in an unbiased manner and the individual
items must be independent.

Practical Problems on Correlation Analysis

Illustration: 1

If r = 0.6 and N = 64 of a distribution, Find out the probable error.

Solution:

1  r2 1  (0.6 )2
PE = 0.6745 = 0.6745 = 0.6745  0.08 = 0.06
N 64

Illustration: 2

For a given distribution the value of correlation is 0.64 and its probable error is
0.13274. Find the number of items in the series.

Amity Directorate of Distance and Online Education

Forecasting Techniques 103

Solution: Notes
2
1r
P.E. = 0.6745 x Given, PE = 0.13274 Correlation (r) = 0.64
n
n = ?

1  ( 0.64 ) 2
PE = 0.6745 x
n

0.6745(1  0.4096)
 0.13274 =  n 0.13274 = 0.6745 x 0.5904
n

0.3982
 n = 0.13274  n = 2.999  n = 9 (Approx)

Illustration: 3

Compute Karl Pearson’s Co-efficient of Correlation from the following data and calculate its
probable error and interpreter result.
Marks in Accountancy 77 54 27 52 14 35 90 25 56 60
Marks in English 35 58 60 40 50 40 35 56 34 42
Solution:
Calculation of Karl Pearson’s Co-efficient of Correlation
X X – 49 X2 Y y – 45 Y2 Xy
X Y
77 + 28 784 35 – 10 100 – 280
54 +5 25 58 + 13 169 + 65
27 – 22 484 60 + 15 225 – 330
52 +3 9 40 –5 25 – 15
14 – 35 1,225 50 +5 25 – 175
35 – 14 196 40 –5 25 + 70
90 + 41 1,681 35 – 10 100 – 410
25 – 24 576 56 + 11 121 – 264
56 +7 49 34 – 11 121 – 77
60 + 11 121 42 –3 9 – 33

x = 490 x2 = 5,150 y = 450 y2 = 920    xy = – 1,449

 xy 1449
r = 2 2 =  0.666
x  y 5150  920

1  r2 1  (0.666)2 0.3753
PE = 0.6745 = 0.6745 =  0.119
N 10 3.162

Amity Directorate of Distance and Online Education

104 Quantitative Techniques in Management

Notes Illustration: 4

The following data relate to age of employees and the number of days they were reported
sick in a month.
Age 30 32 35 40 48 50 52 55 57 61
Sick days 1 0 2 5 2 4 6 5 7 8
Calculate Karl Pearson’s Co-efficient of Correlation
Solution:
Calculation of Karl Pearson’s Co-efficient of Correlation
Age x– x X2 Sick days y– y y2 xy
x x y y
30 – 16 256 1 –3 9 + 48
32 – 14 196 0 –4 16 + 56
35 – 11 121 2 –2 4 + 22
40 –6 36 5 +1 1 –6
48 +2 4 2 –2 4 –4
50 +4 16 4 0 0 0
52 +6 36 6 +2 4 + 12
55 +9 81 5 +1 1 +9
57 + 11 121 7 +3 9 + 33
61 + 15 225 8 +4 16 + 60
 = 460 x2 = 1092 y = 40 y2 = 64 xy = 230

460
x =  46
10
40
y = 4
10
 xy 230 230
r =     0.87
2 2 1092  64 263.36
x  y
There is a high degree of positive Correlation between age and number of days reported sick.

Illustration: 5

Calculate Karl Pearson’s Co-efficient of Correlation from the following data.

Marks in Economics 48 35 17 23 47
Marks in Banking 45 20 40 25 45

Amity Directorate of Distance and Online Education

Forecasting Techniques 105

Solution: Notes
Calculation of Co-efficient of Correlation
X X – 34 x2 Y y – 35 y2 XY
x
48 + 14 196 45 + 10 100 + 140
35 +1 1 20 – 15 225 – 15
17 – 17 289 40 +5 25 – 85
23 – 11 121 25 – 10 100 + 110
47 + 13 169 45 + 10 100 + 130
x = 170 x2 = 776 y = 175 y2 = 550 xy = 280

 x 170  y 175
x =   34 y =   35
N 5 N 5
 xy 280 280
r = 2 2 = =  0.429
x  y 776  550 653.299

Illustration: 6

From the following data of six commodities calculate Karl Pearson’s Co-efficient of correlation
between sales and price per kg sold.
Commodities Sales (Kgs) Total Price
Rice 100 8,000
Wheat 120 10,000
Sugar 150 13,000
Gram 100 10,000
Oil 160 19,200
Ghee 170 30,940
Solution:
In order to simplify the calculations the sales column is divided by 10 and price column is
divided by 1000.
Calculate of Coefficient of Correlation
X x = x – 15 x2 Y y = y – 15 y2 xy
10 –5 25 8 –7 49 39
12 –3 9 10 –5 25 15
15 0 0 13 –2 4 0
10 –5 25 10 –5 25 25
16 +1 1 19.2 4.2 17.64 4.2
17 +2 30.94 15.94 15.94 254.08 31.88
x = – 10 x2 = 64 y = 1.14 y2 = 374.72 xy = 111.8

Amity Directorate of Distance and Online Education

106 Quantitative Techniques in Management

Notes
( x )( y)
xy  ( 10)(114
. )
n 111.08 
6
r = ( x ) 2 ( y ) 2 =
x 2  y 2  ( 10)2 . )2
(114
n n 64  374.72 
6 6

112.98 112.98
= =  10.84
4733  374.50 133.14

As there is a high positive correlation between sales and total price it means when
Sales is high, the total Price will also be high.

Illustration: 7

From the data given below, compute Karl Pearson’s co-efficient of correlation.
X series Y series
Number of items 15 15
Arithmetic mean 25 18
Square of deviation from Arithmetic mean 136 138
Summation of products of deviations from the arithmetic means of X and Y-122.
Solution:
 x = 25,  x2 = 136  xy = 122
 y = 18  y2 = 138 n = 15

( x )( y)  25 x 18 
xy  122   
n  15 
r = ( x )2 ( y) 2 = (25)2 (18)2
x 2  y 2  136  138 
n n 15 15

122  30 92 92
= = (9.71)(10.78)  104.67 = 0.8789
136  41.66 138  21.6
 Karl Pearson’s co-efficient of correlation is 0.8789

Illustration: 8

Calculate Karl Pearsons co-efficient of correlation for the following data regarding
price and demand of a certain commodity.
Price (in `) 21 22 23 24 25 26 27 28 29
Demand (in ‘000 units) 20 19 19 17 17 16 16 15 14
Solution:

Price (X) x = X – A x2 Demand(y) y (Y A) y2 xy

A = 25 A = 17
21 –4 16 20 3 9 –12
22 –3 9 19 2 4 –6

Amity Directorate of Distance and Online Education

Forecasting Techniques 107

23 –2 4 19 2 4 –4 Notes
24 –1 1 17 0 0 0
25 0 0 17 0 0 0
26 1 1 16 –1 1 –1
27 2 4 16 –1 1 –2
28 3 9 15 –2 4 –6
29 4 16 14 –3 9 –12
x = 0 60 y = 0 32

 Karl Pearson co-efficient correlation

( x )( y )
xy 
n 43  0  43
r = 2 ( x ) 2 2( y )2 = = 60 x 32
x  y  60  0 32  0
N N

 43  43
= = = –0.981
1920 43 .81

Illustration: 9

Calculate Karl Pearson’s coefficient of correlation between percentage of pass and

failures from the following data. Also obtain probable error:
No. of Students: 80 60 90 70 50 40
No. of Students passed: 48 30 45 56 45 30
Solution:

% of pass % of fail x = x – a x2 y = y – b y2 x y
(x) (y) a = 50 b = 50

60 40 10 100 -10 100 -100

50 50 0 0 0 0 0
50 50 0 0 0 0 0
80 20 30 900 -30 900 -900
90 10 40 1,600 -40 1,600 -1,600
75 25 25 625 -25 625 -625

405 195 105 3,225 -105 3,225 -3,225

405 195
X = 6
= 67.5 =
6
= 32.5

N .xy  x . y
r =
Nx  x ) 2
2
Ny 2  ( y ) 2

6( 3,225)  (105 )( 105 )

=
6(3,225 )  (105) 2 6(3,225 )  ( 105 )2
Amity Directorate of Distance and Online Education
108 Quantitative Techniques in Management

Notes
 19,350  11,025  8,325  8,325
= 19,350  11,025 19,350  11,025 = 2 = = –1
( 8,325) 8,325

PE = 0.6745
1  r2
=

0.6745 1  ( 1)2  =
0.6745 x 0
= 0
n 6 6

3.7 Rank Correlation

When statistical series in which the variables under study are not capable of
quantitative measurement but can be arranged in serial order, in such situation pearson’s
correlation co-efficient cannot be used in such case Spearman Rank correlation can be
used.
This is a measure of correlation which is used when quantitative measures for certain
factors such as evaluation of leadership ability, judgement of female beauty etc cannot
be fixed, but the individual in the group can be arranged in order thereby obtaining for
each individual a number indicating his or her rank in the group. It is obtained by applying
the following formula:

6  D2
rn = 1
N ( N 2  1)
Where, r = Rank correlation
D = Difference of rank between paired items in two series
N = Total number of observation

Features of Spearman’s Rank Correlation

The features of Spearman’s rank correlation are as follows:

1) The sum of the differences of ranks between two variables shall be zero, i.e.
 d = 0.
2) Spearman’s Correlation co-efficient is distribution – free or non-parametric
because no strict assumptions are made about the form of population from which
sample observations are drawn.
3) The Spearman’s Correlation Co-efficient is nothing but Karl Pearson’s
Correlation Co-efficient between the ranks. Hence it can be interpreted in the
same manner as karl Pearson’s correlation co-efficient.

How is rank Correlation calculated in case of equal ranks?

When equal ranks the assigned to some entries an adjustment in the usual formula
for calculating the rank coefficient of correlation is made. The adjustment consists of adding
1/12 (m2 – m) to the value of  d 2 , where ‘m’ Stands for the number of items whose
ranks are common. If there are more than one such group of items with common rank,
this value is added as many times, the number of such groups. The formula can this be
written

RS
6  d2 
1
(m3  m) 
1
(m3  m) 
UV
rn = 1 T 12 12 W
N3  N
Amity Directorate of Distance and Online Education
Forecasting Techniques 109

Interpretation of Rank Correlation Coefficient (R) Notes

a) The value of rank correlation coefficient, R ranges from -1 to +1.
b) If R = +1, then there is complete agreement in the order of the ranks and the
ranks are in the same direction.
c) If R = -1, then there is complete agreement in the order of the ranks and the
ranks are in the opposite direction.
d) If R = 0, then there is no correlation.

Procedure for Computing Rank Correlation Coefficient (R)

a) Problems where actual rank are given-

i) Calculate the difference ‘D’ of two Ranks i.e., R 1

 R2 .
ii) Square the difference and calculate the sum of the difference i.e.,  d 2
iii) Substitute the values obtained in the formula.
b) Problems where Ranks are not given-
If the ranks are not given, then need to assign ranks to the data series. The lowest
value in the series can be assigned rank 1 or the highest value in the series can be
assigned rank 1. To follow the same scheme of ranking for the other series. Then calculate
the rank correlation coefficient in similar way as the ranks are given.

Merits of Spearman’s Rank Correlation

The advantages of rank correlation are as follows:

a) This method is simpler to understand and easier to apply compared to karl
pearson’s correlation method.
b) This method is useful where we can give the ranks and not the actual data.
c) This method is to use where the initial data in the form of ranks.

Limitation Spearman’s Correlation

The disadvantages of rank correlation are as follows:
a) Cannot be used for finding out correlation in a grouped frequency distribution.
b) This method should be applied where N exceeds 30.

Practical Problems on Rank Correlation Analysis

Illustration: 1

From the following data calculate co-efficient of rank correlation between X and Y.
X: 36 56 20 65 42 33 44 50 15 60
Y: 50 35 70 25 58 75 60 45 80 38
Solution:
Calculation of co-efficient of Rank Correlation
X R1 Y R d= R -R d2
2 1 2
36 7 50 6 1 1
56 3 35 9 -6 36
20 9 70 3 6 36
Amity Directorate of Distance and Online Education
110 Quantitative Techniques in Management

Notes 65 1 25 10 -9 81
42 6 58 5 1 1
33 8 75 2 6 36
44 5 60 4 1 1
50 4 45 7 -3 9
15 10 80 1 9 81
60 2 38 8 -6 36
d=0 d2=318

6d2 6x318 1908 1908

Rk = 1- = 1- = 1- = 1- = 1-1.92727
n n3 (10)3  10 1000 10 990
= -0.92727 or -0.93

Illustration: 2

For the following data, calculate the coefficient of rank correlation:

X: 80 91 99 71 61 81 70 59
Y: 123 135 154 110 105 134 121 106
Solution:

X Y Rank Rank d = x–y d2

for x for y

80 123 4 4 0 0
91 135 2 2 0 0
99 154 1 1 0 0
71 110 5 6 –1 1
61 105 7 8 –1 1
81 134 3 3 0 0
70 121 6 5 1 1
59 106 8 7 1 1

n = 4

6d2 6(4) 24
r = 1 – = 1 – = 1 –
2
n(n  1) 8(82  1) 8(63)

24
= 1 – = 1 – 0.0476
504
r = 0.9524

Illustration: 3

Calculate co-efficient of Rank correlation from the following data:

X: 60 34 40 50 45 41 22 43 42
Y: 73 32 34 40 45 33 12 30 36

Amity Directorate of Distance and Online Education

Forecasting Techniques 111

Solution: Notes
Computation of Rank Correlation co-efficient
X Y R1 R2 d = R1 – R2 d2
60 75 1 1 0 0
34 32 8 7 1 1
40 34 7 5 2 4
50 40 2 3 –1 1
45 45 3 2 1 1
41 33 6 6 0 0
22 12 9 9 0 0
43 30 4 8 –4 16
42 36 5 4 1 1
d2 = 24

1  6d2 6 x 24 144
r = = 1 1 = 1 – 02 = 0.8
3
n n 93  9 720

Illustration: 4

Calculate the co-efficient of rank correlation for the following data of marks obtained by 9
students in Accountancy and Statistics at an Examination.

Marks in Accountancy Marks in Statistics

80 30
55 8
60 17
36 13
60 25
25 12
90 13
15 2
60 5
Solution:
Marks in Marks in Rank Rank d= d2
Accountancy (X) Statistics(Y) Rx Ry Rx–Ry
80 30 2 1 1 1
55 8 6 7 -1 1
60 17 4 3 1 1
36 13 7 4.5 2.5 6.25
60 25 4 2 2 4
25 12 8 6 2 4
90 13 1 4.5 - 3.5 12.25

Amity Directorate of Distance and Online Education

112 Quantitative Techniques in Management

Notes 15 2 9 9 0 0
60 5 4 8 - 4 16
d2=45.5

 3 3
(m1  m) (m2  m) 
6 d2  
 12 12 
Rank correlation co-efficient = r 1   
s
n3  n

Where
 (33  3) (23  2)  m  no. of times each
6 45.5  
 12 12 
1    Valuerepeats
3
9  9
n  totalno. of values

6(45.5  2  0.5) 6(48)

= 1  1 = 1 - 0.4  rs = 0.6
729  9 720

Illustration: 5

Calculate the rank correlation between the length of service and order of merit from the
following data:
Employee: A B C D E F G H I J K L
Yrs of service: 5 2 10 8 6 4 12 2 7 5 9 8
Order of Merit: 6 12 1 9 8 5 2 10 3 7 4 11
(Efficiency)
Solution:
Years of service should be ranked, whereas the order of merit not.

Yrs of Service X Rx Merit Ry Rx - Ry = d d2

5 8.5 6 + 2.5 6.25

2 11.5 12 - 0.5 0.25
10 2 1 +1 1
8 4.5 9 -4.5 20.25
6 7 8 -1 1.00
4 10 5 5 25.00
12 1 2 -1 1
2 11.5 10 +1.5 2.25
7 6 3 3 9
5 8.5 7 1.5 2.25
9 3 4 -1 1
8 4.5 11 -6.5 42.25

d = 0  d2 = 111.5

Tied Ranks - (I) In X series (a) 8 th and 9th and (b) 11th and 12th and (c) 4th and 5th

Amity Directorate of Distance and Online Education

Forecasting Techniques 113

1 1 1 Notes
6[d2  (23  2)  (23  2)  (23  2)]
rs  1  12 12 12
n3  n

6[111.5  0.5  0.5  0.5] 6(113) 678

 1  1 = 1 = 1 - 0.395 = 0.605
123  12 1716 1716

3.8 Regression Analysis

Regression is the measure of the average relationship between two or more variables
in terms of the original units of the data.
The term regression analysis refers to the methods by which estimates are made
of the values of a variable from a knowledge of the values of one or more other variables
and to the measurement of the errors involved in this estimation process.
Regression Analysis is mathematical measure of average relationship between two
or more variables. Regression analysis is a statistical tool used in prediction of value of
unknown variable from known variable. Regression Analysis is a very powerful tool in the
field of statistical analysis in predicting the value of one variable, given the value of another
variable, when those variables are related to each other.

Regression Co-efficient

The rate of change of Variable for unit change in the other variable is called the
regression co-efficient of former on the latter. Since there are two regression lines there
are two regression co-efficients. The rate of change of x for unit change in y is called
regression coefficient of x on y. It is the Co-efficient of y in the regression equation. When
it is in the form of x = n + by. This is denoted by bxy

x
bxy = r
y

x
Similarly, regression Co-efficient of y on x is denoted by byx = r
y

Assumptions in Regression Analysis

a) Existence of actual linear relationship.
b) The regression analysis is used to estimate the values within the range for which
it is valid.
c) The relationship between the dependent and independent variables remains the
same till the regression equation is calculated.
d) The dependent variable takes any random value but the values of the independent
variables are fixed.
e) In regression, we have only one dependant variable in our estimating equation.
However, we can use more than one independent variable.

Utilities of Regression Analysis

(1) Regression analysis is used in all those fields where two or more related
variables are having the tendency to go back to the average.

Amity Directorate of Distance and Online Education

114 Quantitative Techniques in Management

Notes (2) It predicts the values of dependent variables from the known values of
independent variables.
(3) It is used in Statistical estimation of demand curve, supply curve, Production
function, Cost function, consumption function etc.
(4) To study correlation with the help of regression.

Regression Lines

If we take two variables X and Y we shall have two regression lines as the regression
of X on Y and the regression of Y on X. The regression line of Y on X gives the most
probable values of Y for given value of X and the regression line of X on Y gives the most
probable values of X for given values of Y. However, When there is either perfect positive
or negative correlation between the two variables, the regression lines will coincide, i.e.,
we will have only one line.
The Regression Line is the line that best fits the data, such that the overall distance
from the line to the points (variable values) plotted on a graph is the smallest. In other
words, a line used to minimize the squared deviations of predictions is called as the
regression line.
The regression line gives the average relationship between the two variables in
mathematical form. The Regression would have the following properties:
a)  (Y – Yc) = 0 and

b)  (Y – Yc)2 = Minimum

Regression Lines Signify

The farther the two regression lines from each other, the lesser is the degree of
correlation and the nearer the two regression lines to each other, the higher is the degree
of correlation. If the variables are independent, r is zero and the lines of regression right
angles, i.e., parallel to ox and oy.
For two variables X and Y, there are always two lines of regression –
Regression line of X on Y:
gives the best estimate for the value of X for any specific given values of Y?
X = a + bY
a = X - intercept
b = Slope of the line
X = Dependent variable
Y = Independent variable
For two variables X and Y, there are always two lines of regression –
Regression line of Y on X:
gives the best estimate for the value of Y for any specific given values of X
Y = a + bx
a = Y - intercept
b = Slope of the line

Amity Directorate of Distance and Online Education

Forecasting Techniques 115

Y = Dependent variable Notes

x = Independent variable

Properties of the Regression Co-efficients

The co-efficient of correlation is geometric mean of the two regression co-efficients.

r  byx  bxy
If byx is positive than bxy should also be positive and vice versa.
If one regression coefficient is greater than one the other must be less than one.
The coefficient of correlation will have the same sign as that our regression coefficient.
Arithmetic mean of byx and bxy is equal to or greater than co-efficient of correlation.
byx + bxy / 2  r Regression coefficient are independent of origin but not of scale.

Standard Error of Estimate

Standard Error of Estimate is the measure of variation around the computed

regression line.
Standard error of estimate (SE) of Y measure the variability of the observed values
of Y around the regression line.
Standard error of estimate gives us a measure about the line of regression. of the
scatter of the observations about the line of regression.

Standard Error of Estimate of Y on X is: S.E. of Yon X (SExy) =  (Y –

Ye )2 / n-2
Y = Observed value of y
Ye = Estimated values from the estimated equation that correspond to each y
value
e = The error term (Y – Y e )
n = Number of observation in sample.

The convenient formula: (SExy) =  Y2 _ a  Y _ b  YX / n – 2

X = Value of independent variable.
Y = Value of dependent variable.
a = Y intercept.
b = Slope of estimating equation.
n = Number of data points.
• Regression Coefficient of X on Y: The regression coefficient of X on Y is
represented by the symbol bxy that measures the change in X for the unit
change in Y. Symbolically, it can be represented as: The bxy can be obtained
by using the following formula when the deviations are taken from the actual
means of X and Y: When the deviations are obtained from the assumed mean,
the following formula is used:

Amity Directorate of Distance and Online Education

116 Quantitative Techniques in Management

Notes • Regression Coefficient of Y on X: The symbol byx is used that measures

the change in Y corresponding to the unit change in X. Symbolically, it can
be represented as:
In case, the deviations are taken from the actual means; the following formula is
used:
The byx can be calculated by using the following formula when the deviations are
taken from the assumed means:
The Regression Coefficient is also called as a slope coefficient because it determines
the slope of the line i.e., the change in the independent variable for the unit change in
the independent variable.

Estimation of Regression Line

Regression equations are algebraic expressions of the regression lines. Since there
are two regression lines, There are two regression equations – The regression equation
of X on y is used to describe the variations in the values of X for given changes in y and
the regression equation of y on x is used to describe the variation in the values of y for
given changes in x
The two Regression Equations
The two regression equations are as follows:
(1) Regression equation of x on y
Y = a + bx
The Values of a and b obtained by solving the two simultaneous equation
y = Na + bx
xy = ax + bx2
(2) Regression equation of x on y
X = a + by
The values of a and b are obtained by solving the two simultaneous equations
x = Na + by
y = ay + by2

Practical Problems Regression Analysis

Illustration: 1

Compute two regression co-efficient When r = 0.8,  x = 5,  y = 7.

x 5
bxy = r y = 0.8  = 0.57
7
y 7
byx = r = 0.8   112
.
x 5

Amity Directorate of Distance and Online Education

Forecasting Techniques 117

Illustration: 2 Notes
State the relationship between Correlation co-efficient of two variables and their
regression co-efficient of two variables and their regression co-efficients.
Regression Co-efficient of x on y =
x
bxy = r y
Regression Co-efficient of y on x
y
byx = r
x
1 y
bxy byx = r r
y x = r2  r

 r= bxy. byx
It can be Said that correlation Co-efficient is the geometric mean of regression
Co-efficients.

Illustration: 3

Obtain the two regression equations from the following

X – Series Y – Series
Mean 20 25
Variance 4 9
Coefficient of correlation = 0.75
Solution:
The regression equation of x on y is
x
(x  x ) = r  ( y  y)
y

(x - 20) = 0.75 
2
3
by  25 g = 0.50 y 
2
3
by  25 g
= 0.50y + 20 – 12.5 = 0.50y + 7.5
The regression equation of y on x is

y 3
( y  y ) = r  ( x  x ) (y – 25) = 0.75  ( x  20)
x 2
= 1.125n + 25 – 22.5 y = 1.125n + 2.5
The equations are
x = 0.5y + 7.5
x = 1.125x + 2.5

Amity Directorate of Distance and Online Education

118 Quantitative Techniques in Management

Notes Illustration: 4
You are given the following information about advertisement and Sales of a Company
Mean 10 90
Variance 9 16
Correlation Co-efficient = 08
(1) Calculate the two regression lines
(2) Find the likely Sales When advertising expenses are ` 15 lakhs and
(3) What should be the advertisement expenses
If the Company Wants to attain a Sales target of ` 190 Lakhs.
Solution:
Regression equation of x on y
x 3
( x  x ) = r  ( y  y ) x – 10 = 0.8 ( y  90 )
y 4
x – 10 = 0.6y – 54 x = 0.6y – 44
The advertisement expenses for Sales Target of ` 190 lakhs will be
X = 0.6  190 – 44 = ` 70 Lakhs
Regression equation of y on x:
y 4
( y  y ) = r  ( x  x ) y – 90 = 0.8  ( x  10 )
x 3
x – 90 = 1.0667x – 10667 y = 1.0667x + 79.333
Sales When advt expenses is ` 15 Lakhs
Y = 1.0667  15 = 79.333 = 95.33 Lakhs

Illustration: 5

Obtain the two regression equations by taking deviations from actual mean of x and
y Series
X 2 4 6 8 10
Y 5 7 9 8 11
Solution:
Calculation of Regression equations
X X–6=x X2 Y y–8=y Y2 Xy
2 –4 16 5 –3 9 12
4 –2 4 7 –1 1 2
6 0 0 9 1 1 0
8 2 4 8 0 0 0
10 4 16 11 3 9 12
x = 30 x2 = 40 y = 40 =0 y2 = 20 xy = 26
Amity Directorate of Distance and Online Education
Forecasting Techniques 119

 x 30  y 40 Notes
x = N  5 6 y =
N

5
8

Regression equation of x on y:

 xy 26
(x  x ) = b( y  y) b= =  1.3
 y2 20
(x – 6) = 1.3 (y – 8) x – 6 = 1.3y – 10.4 x = 1.3y – 4.4
Regression equation of y on x:

 xy
( y  y) = b(x  x ) b=
 x2

26
=  0.65 (y – 8) = 0.65 (x – 6) y – 8 = 0.65x – 3.9
40
y = 0.65x + 4.1

Illustration: 6

To study the relationship between expenses on accommodation ‘X’ and expenditure

on food and entertainment ‘Y’ on enquiry into 50 families gave the following results
x = 8500, y = 9600, x = 60, y = 20, r = + 0.6
Estimate the expenditure on food and entertainment When expenditure on
accommodation is ` 200.
Solution:
Let x services represent expenditure on accommodation and y Series entertainment
Regression equation of y on x:

y 20
( y  y ) = r  ( x  x ) (y - 192) = 0.6  ( x  170 )
x 60
y – 192 = 0.2x – 34 y = 0.2x + 158

 x 8500
x = N  50  Rs.170

 y 9600
y =   Rs.192
N 50
Given x = 200, y = ?
Y = 0.2  200 + 158 = 198

Illustration: 7

The following table shows the exports of New Cotton and the imports of manufactured
goods into India for 7 years
Exports (in Crores of `) 42 44 58 55 89 98 60
Imports (in crores of `) 56 49 53 58 67 76 58

Amity Directorate of Distance and Online Education

120 Quantitative Techniques in Management

Notes Obtain the two regression equations and estimate the imports when exports in a
particular year were to the value of ` 70 crores.
Solution:
‘X’ Series represents Export ‘Y’ Series represents Import
Calculation of Regression equations

X dx=x–60 dx2 Y dy = y – 60 dy2 dxdy

42 – 18 324 56 –4 16 72

44 – 16 256 49 – 11 121 176

58 –2 4 53 –7 49 14

55 –5 25 58 –2 4 10

89 29 841 67 7 49 203

98 38 1444 76 16 256 608

60 0 0 58 –2 4 0

x = 446 dx2 = 2894 y = 417dy = – 3 dy2=499 dxdy=1083

 x 446
x = N  7  63.714

 y 417
y =   59.571
N 7
Regression equation of x on y:

x  x = bxy (y – y )

N  dxdy  (  dx )(  dy)
bxy =
N  dy 2  (  dy)2
7  1083  26  3 7581  78 7659
= 2 = =  2.1983
7  499  ( 3) 3484 3484

(x – 63.714) = 2.1983 (y – 59.571)

x – 63.714 = 2.1983y – 130.9549
x = 2.1983y – 67.2409
Regression equation of y on x

y  y = byx ( x  x )

Amity Directorate of Distance and Online Education

Forecasting Techniques 121

N  dxdy  (  dx )(  dy) Notes

byx =
N  dx 2  (  dx )2

7  1083  26  3 7659
= 2 =  0.3911
7  2894  (26) 19582
y – 59.571 = 0.3911(x – 63.714)
y – 59.571 = 0.3911x – 24.9185
y = 0.3911x + 34.6525
Calculate of y When x = 70
y = 0.3911  70 + 34.6525 = 62.0295 crores

Illustration: 8

The following data gives the demand (x) and Supply (y) in a market for 11 days
X 61 72 73 73 63 84 80 66 76 74 72
Y 40 52 59 53 61 58 56 42 58 50 50
Solution:
Calculation of Regression equation
X dx = x–72 dx2 Y dx = y – 52 dy2 dxdy
61 – 11 121 40 – 12 144 132
72 0 0 52 0 0 0
73 1 1 59 7 49 7
73 1 1 53 1 1 1
63 –9 81 61 9 81 – 81
84 12 144 58 6 36 72
80 8 64 56 4 16 32
66 –6 36 42 – 10 100 60
76 4 16 58 6 36 24
74 2 4 50 –2 4 –4
72 0 0 50 –2 4 0
x = 794 dx2 = 468 y = 579 dy = 7 dy2=471 dxdy=243
Regression equation of x on y:
N  dxdy  (  dx )(  dy)
bxy =
N  dy 2  (  dy)2
11  243  2  7 2659
= =  0.5181
11  471  (7)2 5132
x – 72.18 = 0.5181 (y – 52.64)
x – 72.18 = 0.5181 y – 27.2738
x = 0.5181y + 44.9062

Amity Directorate of Distance and Online Education

122 Quantitative Techniques in Management

Notes Regression equation of y on x:

y  y = bys ( x  x )

N  dxdy  (  dx )(  dy) 11  243  2  7 2659

byx = 2 2 = 2 =  0.5169
N  dx  (  dx ) 11  468  (2) 5144
y – 52.64 = 0.5169 (x – 72.18)
y – 52.64 = 0.5169x – 37.3098
y = 0.5169x + 15.73

r= byx  bxy = 0.5169  0.5181 = + 0.5175

 x 794
x = N  11  72.18

 y 579
y =   52.64
N 11

Illustration: 9

The following data relate to the age of husbands and wives. Obtain the two regression
equations and determine the most likely age of husband when age of wife is 25 years
Age of Husbands 25 28 30 32 35 36 38 39 42 55
Age of Wives 20 26 29 30 25 18 26 35 35 46
Solution:
Age of Husband is represented by X Series and age of Wives is represented by Y Series

X X = x – 36 X2 Y Y = y – 29 Y2 Xy
25 – 11 121 20 –9 81 99
28 –8 64 26 –3 9 24
30 –6 36 29 0 0 0
32 –4 16 30 1 1 –4
35 –1 1 25 –4 16 4
36 0 0 18 – 11 121 0
38 2 4 26 –3 9 –6
39 3 9 35 6 36 18
42 6 36 35 6 36 36
55 19 361 46 17 289 323
x = 360 x2 = 648 y = 290 y2 = 598 xy = 494

 x 360
x = N  10  36 years
 y 290
y =   29 years
N 10

Amity Directorate of Distance and Online Education

Forecasting Techniques 123

Regression equation of X on Y Notes

x  x = b( y  y)
 xy 494
b= 2
  0.8261 x – 36 = 0.8261(y – 29)
y 598
x – 36 = 0.8261y – 23.9569 x = 0.8261 y + 12.0431
When Age of Wife is 25 Age of husband will be
x = 0.8261 x 25 + 12.043 = 32.6156 or = 33 years
Regression equation of y on x:
y  y = b(x  x )
 xy 494
b= 2
  0.7623 y – 29 = 0.7623 (x – 36)
x 648
y – 29 = 0.7623x – 27.4428 y = 0.7623x + 1.5572

3.9 Time Series Analysis

The observed values of the variable studied, such as the price of a commodity are
results of various influences. Discovering and measuring the effects of these influences
are the primary purposes of a time series analysis. Although the effects cannot always
be determined exactly, been made over a sufficiently long period.
Time series analysis is done primarily for the purpose of making forecasts for future
and also for the purpose of evaluating past performances. An economist or a business
man is very naturally interested in estimating the future figure of national income,
population, prices and wages etc. So the success or failure of a businessman depends
to a large extent on the accuracy of this future forecasts.

Meaning of Time Series

Time series analysis are basic to understanding past behaviour, evaluating current
accomplishments, planning for future operations and comparing different time series. Thus
a series of successive observations of the same phenomenon over a period of time are
called “time series”.

Definitions

According to Patterson, “A time series consists of statistical data which are

collected, recorded or observed over successive increments.”
“A set of data depending on the time is called time series”. – Kenny and keeping
“A time series is a set of statistical observations arranged in chronological order”.
– Morris Hamburg
According to Croxton and Cowden, “A time series consists of data arranged
chronologically.”
“A time series is a set of observations taken at specified time, usually at ‘equal
intervals’. Mathematically, a time series is defined by the values y1, y2 ...... of a variable

Amity Directorate of Distance and Online Education

124 Quantitative Techniques in Management

Notes Y (temperature, closing price of share, etc.) at the time t1, t2 .... Thus, Y is a function
of t, symbolised by y = F(t).” – Spiegel
“A time series may be defined as a collection of magnetite belonging to different
time periods, of some variable or different time periods, of some variable or composite
of variables, such as production of steel, per capita income, gross national product, price
of tobacco or index of industrial production.” – Ya–Lun–Chou

Uses of Time Series

The uses of time series is of well significance to the economist, scientist, sociologist,
biologist, researcher and businessman etc. The following uses are:
(i) It helps to understand past behaviour.
(ii) It helps in evaluating current accomplishment.
(iii) It helps in planing future operations.
(iv) It also helps in comparing the actual performance.
(v) It helps to study the factor which influence the changes in economic activities
and predict the future variations in them with certain limitations.

3.10 Trend Analysis

The analysis of time series consists of a description of a component movement
present. To motivate procedures involved in such description, following figure which refers
to an ideal time series.

(i) Long term trend (i) Long term trend and

cyclical movement

(i) Long term trend movement

cyclical and seasonal movement

Fig. (i) Shows a graph of a long term or secular trend line. Fig. (ii) shows this long
term trend line with a superimposed cyclical movement. fig. (iii) shows a superposition
of a seasonal movement on the graph of figure (ii).
Amity Directorate of Distance and Online Education
Forecasting Techniques 125

The ideas presented above provide us with a possible technique for analysis time Notes
series. So, assume that the time series variable Y is a assume that the time series variable
Y is a product of the variables, T C S and I which produce respectively the trend, cyclical,
seasonal and irregular movements.

Graphs of Time Series

A time series involving a variable Y is represented pictorially by constructing a graph

of Y vs. t. The graph of T professional of the India during the years.

70
65
60
55
Num ber of professional

50
45
(in lakhs)

40
35
30
25
20
15
10

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Year
Fig: IT professional of India source: Department of IT

3.11 Components of Time Series

Components
of
time serices

Long term Seasonal movement

or or
Secular movement Seasonal variation

Cydical movement Irregular

or or
Cyclical variations Ramdom movement

Amity Directorate of Distance and Online Education

126 Quantitative Techniques in Management

Notes Characteristic movement of time series may be classified into four main types, often
called components of a time series. The components of a time series are the various
elements which can be separated from the observed data. The components of time series
are classified as under:

1. Long-term or Secular Analysis

The secular trend is the basic long–term tendency, of a particular activity to grow
or to decline. It indicates the presence of a factors that persist for a considerable duration
factors such as population changes, technological improvements, price level fluctuations
or various conditions that are peculiar to individual industries or establishments. Thus,
the tendency of data to grow or decline over a long period is technically called secular
movement or simple trend. For example in India, the sales of car is showing an increasing
trend and sales of black and white TV is showing declining trend.
“The secular trend is that characteristic of a time series which extends consistently
throughout the entire period of time under consideration”. – Henry L. Alder

2. Cyclical Analysis

Its refer to the long term swing about a trend line or curve. These cycles, as they
are sometimes called may or may not be periodic, i.e., they may or may not follow exactly
similar pattens after equal intervals of time. In business and economic activities,
movements are considered cyclical only if they recur after time intervals of more than a
year. An important example of cyclical movements are the so called business cycles
representing intervals of prosperity, recession, desprsion and recovery.
Up and down movements which are different from seasonal fluctuations, in that they
extend over longer period of time–usually two or more years.” – Linclon L. Choo.

3. Seasonal Analysis

Seasonal variations in general refer to annual periodicity in business as economic

theory, the ideas involved can be extended to include periodicity over any interval of time
such as daily, hourly, weekly etc., depending on the type of data available.
Although the word “seasonal” might seem to imply connection with the seasons of
the year, the term means to include any kind of variation which is of a periodic nature,
and whose repeating cycle are of relatively short duration. Seasonal variations are the
result of such seasonal factors as climatic conditions, holidays, business operations or
human habits. A time series giving the number of visitors per hour at a Bank would show
a seasonal variations since noon hours and pay days are always busy periods. The
borrowing of books at college libraries shows seasonal variations, there is a surprising
increase before each examination period and a sharp decline there after.
“Seasonal variations are variations that occur in regular sequence at specific intervals
of time”. – Edward B. Roessler

4. Irregular or Random Analysis

Its refer to the sporadic motions of time series due to chance events such as floods,
strikes, elections, fires, earthquakes, unusual weather etc. Random variations in a time
series are variations which cannot be accounted for by the secular trend, seasonal
movements or cyclical fluctuations. So, it is ordinarily assumed that such events produces

Amity Directorate of Distance and Online Education

Forecasting Techniques 127

variations lasting only short time, it is conceivable that they may be so intense as to Notes
result in new cyclical or other movements.
“Random or erratic variations are variations which occur in a completely unpredictable
fashion”. – Henry L. Alder.

200
180
160
Production (in lakhs)

140
Seasonal
120
100 Cyclical
Trend
80
60
40
20

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Year

Example - 1

Using the method of Free hand determine the trend of the following data:
Year: 1998 1999 2000 2001 2002 2003 2004 2005
Production: 42 44 48 42 46 50 48 52
(in tonnes)
Solution:
Y

56
X axis 1cm = year
54 Y axis 2 cm = 2 tonce
52
Production

50
48
46 Trend line
44 Actual data

42
40

O X
1998 1999 2000 2001 2002 2003 2004 2005
Years

3.12 Methods of Finding Trend

Estimation of trend values can be achieved in several methods:
(i) Graphic or Free hand Curve Method
(ii) Semi-Average Method

Amity Directorate of Distance and Online Education

128 Quantitative Techniques in Management

Notes (iii) Moving Average Method

(iv) Least Square Method

Graphic or Free hand Curve Method

Which consists of fitting a trend line or curve simply by looking at the graph, can
be used to estimated trend.
The time is shown on the horizontal axis and the value of the variable on the vertical
axis. The fitting of the trend may be straight line or a curved line and it may be done
free hand by scale rules, spline, string or French curves of different shapes. Smooth out
irregularities by drawing a free hand curve through the scatter points.
Merits:
(i) A free hand trend fitting enables an understanding of the character of time series.
(ii) These are its flexibility and simplicity.
(iii) This method only used to describe all types of trends – linear and non linear.
Demerits:
(i) This is depending too much on individual judgment.
(ii) It does not involved any complex mathematical techniques.
(iii) It does not enable us to measure trend in precise quantitative terms.
(iv) Different trend curves could be obtained by different persons for the same data.
It is highly subjective.

Semi–Average Method

The method is used only when the trend is linear or almost linear. For non-linear
trends this method is not applicable. It is used for the calculation of averages, and averages
are affected by extreme values. Thus if there is some very large value or very small value
in the time series, that extreme value should either be omitted or this method should not
be applied.
Merits:
(i) It is simple and easy to understand.
(ii) It can compared with the moving average of the least squares method of
measuring trend.
(iii) This method is objectivity in the sense that it does not depend of the personal
judgment.
(iv) It is also applicable where the trend is linear or approximately linear.
Demerits:
(i) This method is based on the assumption that there is a linear trend which may
not be true.
(ii) It is not suitable when time period represented by average is small.
(iii) The use of arithmetic mean for obtaining semi–average may be questioned
because of limitation of the method.

Illustration - 1

Using the method of semi–average determine the trend of the following data:

Amity Directorate of Distance and Online Education

Forecasting Techniques 129

Year : 1998 1999 2000 2001 2002 2003 2004 2005 Notes
Production: 48 42 60 54 38 66 70 66
Solution:
The number of observations are even i.e., 8. The two middle parts will be 1997 to
2001 and 2002 to 2006.

Year Actual value 4 years total semi–average

and average value
1888 48
48  42  60  54 204
1999 42  51 ...... .... x1
4 4
2000 60
2001 54
38  66  70  66 240
2002 38  61 ...... x2
4 4
2003 66
2004 70
2005 66
Here, the value 51 is plotted against the middle of the first four years i.e., 1998–
2001 and the value 60 is plotted against the middle of the last four years i.e., 2002–2005.
So both the point are joined by a straight line as under.

72
68
64
Production

60 –
X2 = 60
56
Trend line
52
–
X1 = 51 Actual line
48
44
40

O X
1998 1999 2000 2001 2002 2003 2004 2005
Years

Illustration - 2

Calculate trend values from the following data by the method of semi–average:
Year: 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Sales: 30 35 37 24 42 36 27 45 40 42

Amity Directorate of Distance and Online Education

130 Quantitative Techniques in Management

Notes Solution:
Year Sales Averages Annual changes Trend value

1996 30 29.2
1997 35 30
170
1998 37  34 30.8
5
1999 24 31.6
2000 42 0.8 32.4
2001 36 33.2
2002 27 .... 34
2003 45 34.8
2004 40 35.6
2005 42 36.4
Let us have two periods of 5 years each 1996 to 2000 and 2001 to 2005. The averages
for the two periods are:

30  35  39  24  42 170
  34.......... x1
5 5

36  27  45  40  42 190
and   38.......... x2
5 5
The increase of two averages is 38 – 34 = 4 which takes place in 5 years, therefore,
annual change is .... = 0.8
“Graph showing Sales with Trend line”

Y
47
45
43
41
–
39 X2
37 Trend line
Sales

35
33 –
X1 Actual line
31
29
27
25

O X
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Years

Amity Directorate of Distance and Online Education

Forecasting Techniques 131

Illustration - 3 Notes
Draw a trend by the method of semi–average from the following data:
Year: 1998 1999 2000 2001 2002 2003 2004 2005
Sales (‘000 units): 195 100 104 90 95 102 110 116
Also predict the Sales fort he year 2003 from the graph.
Solution:
Year Sales(x) Semi Average
1998 195
489
1999 100 = 122.25 ....... x1
4
2000 104
2001 90
2002 95
423
2003 102 .... = 1.5.75 ..... x2
4
2004 110
2005 116

Graph showing sales “000”units

Y
200
190
180
170
160
150
Sales

140
2003 Sales = 108,00 units
130
120 –
X1
Trend line
110
– Actual line
100 X2
90

X
O
1998 1999 2000 2001 2002 2003 2004 2005
Years

3.13 Moving Average Methods

By using moving averages of appropriate orders, cyclical, seasonal and irregular
patterns may be eliminated thus leaving only the trend movement. Here, trend values can
be obtained by employing arithmetic means of the series except at the two ends of the
series. So moving average consists of a series of arithmetic means calculated from
overlapping groups of successive values of a time series. Moving average based on values

Amity Directorate of Distance and Online Education

132 Quantitative Techniques in Management

Notes covering a fixed time interval, called moving average period. It is shown against the centre
of the period.
The moving average for period ‘t’ is a series of successive averages of values at a
time, starting with 1st, 2nd and 3rd to ‘t’ terms. Here, the first average is the mean of
the 1st to ‘t’ terms, the second is the mean of the ‘t’ terms from 2nd to (t + 1) th terms
and third is the men of the 3rd to (t + 2)th terms and so on.
Hence, the time series values X1, X2, X3 .....
for different time periods, the moving average of period ‘t’ is given by:

1
1st value moving average = (X1 + X2 + .... Xt),
t

1
2nd value moving average = (X2 + X3 + ...... Xt + 1) and 3rd
t

1
value moving average = (X + X4 + ..... Xt + 2)
t 3
(a) Odd period: When the period is odd, if the period ‘t’ of the moving average
is odd the successive value of the moving averages are placed against the middle
value of concerned group of items. If t = 5, the first moving average value is
placed against the middle period i.e. third value and the second moving average
value is placed against the time period four and so on.
(b) Even period: When the period is even, it the period ‘t’ of moving average is
even there are two middle periods and the moving average value is placed
between the two middle terms of the time intervals. In the particular period t
= 4, the first moving average is placed against the middle of second and third
values, the second moving average is placed in between third and fourth values
and so on.
For odd: In the below table (a) have considered 5 years moving averages.

Year Data 5 year 5 year

moving total moving average

1995 50.0
1996 36.5
1997 43.0 212.9 42.6
1998 44.5 201.0 40.2
1999 38.9 197.1 39.4
2000 38.1 192.8 39.6
2001 32.6 190.0 38.0
2002 38.7 192.2 38.4
2003 41.7 187.9 37.6
2004 41.1
2005 33.8
Amity Directorate of Distance and Online Education
Forecasting Techniques 133

Here, the first moving total 219.9 of column 3 is the sum of the 1st through 5th entries Notes
of column 2. The second moving total 201.0 is the sum of the 2nd through 6th entries
in column 2 etc.
In practice, after obtaining the first moving total 212.9, the second moving total is
easily obtained by subtracting 50.0 (1st entry of column 2), the result being 201.0.
Succeeding moving totals are obtained similarly. Dividing each moving total by 5 yields
the required moving average.
In the below table ... have considered 4 years moving averages.
For even
Year Data 4 years 4 years
moving total moving total

1995 50.0
1996 36.5
1997 43.0 174.0 43.5
1998 44.5 162.9 40.7
1999 38.9 164.5 41.1
2000 38.1 154.1 38.5
2001 32.6 148.3 37.1
2002 38.7 151.1 37.8
2003 41.7 154.1 38.5
2004 41.1 155.3 38.8
2005 33.8
Here, the 4 year moving totals are obtained as in past (a), except that 4 entries
of column 2 are added instead of 5. Note that the moving totals are centred between
successive years, unlike part (a). This is always the case when an even number of years.
is taken in the moving average. If we consider that 1996, for example, stands for July
1, 1996, the first 4 year moving total is centred at Jan 1, 1997 or Dec. 31, 1996. The
4 year moving averages are obtained by dividing the 4 year moving totals by 4.
Merits:
(i) It is very simple to understand and easy to calculate as compared to other
methods.
(ii) It can be used in all facts of time series analysis, for the measurement of trends
as well as in connection with seasonal cyclical irregular components.
(iii) It calculate is simple because no higher degree mathematical calculations.
(iv) It considers all the values in the series. The extreme values are included in the
process of determining averages.
(v) It is a objective method. No personal judgment like freehand method.
Demerits:
(i) In this method, that data at the beginning and end of a series are lost.
(ii) It may generate cycles or other movements which were not present in the original
data.

Amity Directorate of Distance and Online Education

134 Quantitative Techniques in Management

Notes (iii) Its are strongly affected by extreme values of items. They are said to be sensitive
to.... movement in the data.
(iv) It does not establish functional relationship between the period and the value of
variables.
(v) It cannot determine irregular variations completely.

Illustration - 1

Find trend values by three yearly moving average method.

Year: 1999 2000 2001 2002 2003 2004 2005
Production: 112 138 146 154 170 183 190
(‘000’ in unit)
Solution:

I II III IV V
Year Production Column of 3 Year 3 Year
differences moving total moving average
1999 112
396
2000 138 154 – 112 = 42 .....396 = 132
3
438
2001 146 170 – 138 = 32 ..... 438 = 146
3

470
2002 154 183 – 146 = 37 .... 470 = 157
3

507
2003 170 190 – 154 = 36 ..... 507 = 169
3

543
2004 183 .... 543 = 181
3
2005 190
Calculating total production of first three years which is (112 + 138 + 146 = 396). This
total is placed in column (IV) of three yearly moving total before the middle year. 2000 and in
column (V) the three year moving average 396/3 = 132 is placed before the year 2000 as shown
in the above table. And find the second moving total we find the difference of the production
of a year 2002 and 1999 which is (154 – 112 = 42). This is written in column (III) in front of
the year 2000 and the second moving total is (396 + 42 = 438). Such way it can complete
columns (III) and (IV) of the above table.

Illustration - 2

Find trend values from the following data using three yearly moving averages and show
the trend line on the graph.
Year Price (`) Year Price (`)
1994 52 2000 75

Amity Directorate of Distance and Online Education

Forecasting Techniques 135

1995 65 2001 70 Notes

1996 58 2002 64
1997 63 2003 78
1998 66 2004 80
1999 72 2005 73
Solution:
Computation of Trend Values

Year Price (`) 3 yearly moving 3 yearly

total moving average
1994 52 –
1995 65 175 58.33
1996 58 186 62.00
1997 63 187 62.33
1998 66 201 67.00
1999 72 213 71.00
2000 75 217 72.33
2001 70 209 69.67
2002 64 212 70.67
2003 78 222 74.00
2004 80 231 77.00
2005 73 –

80
Scale :
X axis : 1cm = 1 year
77
Y axis : 1cm = 3 Rs.

74

71

68
Price

65

62
Trend line
59 Actual line

56

53

50

X
O
94 95 96 97 98 99 00 01 02 03 04 05
Years

Amity Directorate of Distance and Online Education

136 Quantitative Techniques in Management

Notes Illustration - 3

Using a 3 yearly moving averages determine the trend values.

Year: 1995 1996 1997 1998 1999 2000 2001 2002
Production: 21 22 23 24 25 26 27 26
(in ‘000 units)
Solution:

Years Production 3 yearly 3 yearly

moving (in 000 units) Moving total average

1995 21 - -
1996 22 66 22
1997 23 69 23
1998 24 72 24
1999 25 75 25
2000 26 78 26
2001 27 79 26.33
2002 26 - -

Y
Graphical Representation

28
X axis 1 year = 1 cm
27 Y area 1 unit = 1 cm

26

25

24

23

22

21

20
X
95 96 97 98 99 00 01 02

Illustration - 4

Find the four yearly moving averages for the following data:
Year: 1991 1992 1993 1994 1995 1996 1997 1998
Values: 30.1 45.4 39.3 41.4 42.2 46.4 46.6 49.2
Amity Directorate of Distance and Online Education
Forecasting Techniques 137

Solution: Notes
Year Values 4 yearly 4 yearly Centeral value
moving Total moving Average
1991 30.1

1992 45.4
156.2 39.05
1993 39.3 40.56
168.3 42.07
1994 41.4 42.195
169.3 42.32
1995 42.2 43.235
176.6 44.15
1996 46.4 45.125
184.4 46.1
1997 46.6

1998 49.2

Illustration - 5
Using four yearly moving averages determine the trend values and also plot the original and
trend values on a graph.

Year Production (1000 units) Year Production (1000 units)

1994 75 2000 96
1985 62 2001 128
1996 76 2002 116
1997 78 2003 76
1998 94 2004 102
1999 84 2005 168
Solution:

Year Production 4yearly 4yearly Centred Year

Moving Average Moving Total Moving Average

1994 75
1995 62 291 72.75 75.125 1996
1996 76 310 77.50 80.25 1997
1997 78 332 83.00 85.50 1998
1998 94 352 88.00 94.25 1999
1999 84 402 100.50 103.25 2000
2000 96 424 106.00 105.00 2001
2001 128 461 104.00 104.75 2002
2002 116 422 105.50 110.50 2003

Amity Directorate of Distance and Online Education

138 Quantitative Techniques in Management

Notes 2003 76 462 115.50

2004 102
2005 168

Graph showing production and trend value for the years 1994-2005
Scale: OX= 1 cm = 1year, OY = 1 cm = 1000 units.

----------- Actual line

----------------------3.y M.A.
7.y.M.A
Scale: 1 cm = 1 year
Sugar Production (In lakhs tonnes) 1 cm = 10 lakhs ons

Actual Line

Trend Line

Years

Illustration - 6

Calculate the trend values by five yearly moving average method and plot the same on a
graph from the following:
Year: 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Sales: 36 42 54 72 66 60 48 75 78 102 93
(’000’units)

Computation of Trend Values

Year Sales 5 years 5 years
moving total moving total

1995 36 –
1996 42 –
1997 54 270 54
1998 72 294 58.8
1999 66 300 60
2000 60 321 64.2
2001 48 327 65.4
2002 75 363 72.5
Amity Directorate of Distance and Online Education
Forecasting Techniques 139

2003 78 396 79.2 Notes

2004 102 –
2005 93 –
Graph showing sales of units for 11 years and 5 yearly trend lines

110

100 Scale :
X axis : 1cm = 1 year
Y axis : 1cm = 10,000 units
90
Sales (1000 units)

80

70

60

50
Actual line
40 5 yearly moving
average line
30

X
O
95 96 97 98 99 00 01 02 03 04 05
Years

3.14 Least Squares Methods

A definition of the term, line of best fit, should give a unique line. Let us proceed by
considering, for each value of X, the absolute value |Y – Yc| of the difference between the actual
 value and the estimated Y value (see below figure). This difference represents the error
committed by using the estimated Y value instead of the actual value. One way to determine
the line of best fit might be to find that line for which the sum of these errors for all the given
values of X, i.e.,  |Y – Yc |, has the smallest possible value. However, the procedure, while
yielding a unique line, leads to mathematical difficulties customarily associated with the
occurrence of absolute values. A better method and one which accomplishes the same aims,
defines the line in such a way that å (Y – Yc)2 has the smallest value. This method, called
the method of least squares, is the one most generally used in statistics for obtaining the
line of best fit.

Amity Directorate of Distance and Online Education

140 Quantitative Techniques in Management

Notes
Y
|Y–Ye|

|Y–Ye| Y
Ye

Ye
Y

X
O X1 X2 X3 X4

The trend line, the which is a “best fit” to a scatter diagram of n points, its employ
a theorem from elementary mathematics.
From the below figure, the equation of any no–vertical line can be written in the form
Y = a + bx,
where, a is the Y–intercept
and b is the slop of the line.
The slope is given b = tan , where  is the angle measured from the positive X
– axis to the line and is positive or negative as  is obtuse.

X
O

Fig: Graph showing of straight line Y–intercept a and angle of inclination a.

To obtain the values of ‘a’ and ‘b’ constants. Instead to following two secondary
equation:
na + bx = Y ..........(i)
ax + b x2 = XY .........(ii)
Here, x represents the number of years or any period for which the data is given.
Mid point in time is taken as the origin, so that negative values in the first half of
the series balance out the positive values in the second half, i.e., Sx = 0.
The period indicating the higher values can be reduced to the minimum symbolically
as under:

Year (Y) Deviation (x) Year (y) Deviation(x)

1999 –3 1998 –7
2000 –2 1999 –5

Amity Directorate of Distance and Online Education

Forecasting Techniques 141

2001 –1 2000 –3 Notes

2002 0 2001 –1
2003 +1 2002 +1
2004 +2 2003 +3
2005 +3 2004 +5
2005 +7
So that x = 0 as the Deviation are calculated from the mean. The sum of deviations
of the actual value from the computed values is equal to zero.
Here, Y = Actual values
Yc = na and xy = bx2
So the value of ‘a’ and ‘b’ determined as -

y xy
a = ; b=
n x2
Finally the trend line, the line of ‘best fit’ will be drawn under the least squares method.
It is the line from which the sum of the squares of the items, measured parallel to the
Y axis, is the least.
So equation represented by
Y = a + bx
least square method for Trend values
Yc = a + bx

y xy
a = ; b =
n x2
Here, x = independent variable (Deviation)
y = Dependent variable on x.
a = The ‘y’ intercept.
b = Indicate slope and signifies the changes in ‘x’ ...constant.
n = Number of observations.
xy = Product of x and y.
x2 = Square of x.
y = Sum of ‘y’ values
xy = Sum of th<e product of x and y values.
x2 = Sum of square value of x.
Merits:
(i) It is objective method which does not variations in the results.
(ii) It is very easy calculation because no higher degree calculation.
(iii) It determines the trend values and also reflects light on the seasonal, cyclical
and irregular variations.
Amity Directorate of Distance and Online Education
142 Quantitative Techniques in Management

Notes (iv) Since it is based on an algebraic method of calculating the trend, it is free from any
bias, subjectivity does not enter into it.
(v) It is a flexible method, the trend values calculated any period and answer should be
accurate.
Demerits:
(i) This method is in appropriate for a very short series and is unnecessary for along one.
(ii) It does not establish functional relationship between the period (x) and the values of
variable ‘y’.
(iii) It is a tedious method and involves more calculations compared to the methods
discussed earlier.
(iv) It can estimate a value only immediate future and not for distant future.

Illustration - 1

Fit a straight line trend by the method of least squares, tabulate the trend values and show
the values on a graph form the following figure of production of sugar factory:
Year: 2007 2008 2009 2010 2011 2012 2013
Production: 80 90 92 83 94 99 92
(‘000’ tons)
Solution:

Computation of trend values

Year Production x x2 xy Trend values (Yc)

1999 80 –3 9 –240 90 + 2 (–3) = 84
2000 90 –2 4 –180 90 + 2 (–2) = 86
2001 92 –1 1 –92 90 + 2 (–1) = 88
2002 83 0 0 0 90 + 2 (0) = 90
2003 94 +1 1 94 90 + 2 (+1) = 92
2004 99 +2 4 198 90 + 2 (+2) = 94
2005 92 +3 9 276 90 + 2 (+3) = 96

n=7 y = 630 x = 0 x2 = 28 xy = 56

The straight line equation, Yc = a + bx

y 630 xy 56
Here, a =   90 and b   2
n 7 x 28

Amity Directorate of Distance and Online Education

Forecasting Techniques 143

Graph showing production for the year 2007 to 2013 Notes

Y
110
Scale :
105 X axis : 1cm = 1 year
Y axis : 1cm = 5000 tonnes
100
Production

95

90
Trend line
85
Actual line
80

X
O
99 00 01 02 03 04 05
Years

Illustration - 2

Fit a straight line trend to the following data taking X as the independent variable and prove
S(Y – Yc) = 0.
X: 2000 2001 2002 2003 2004 2005
Y: 1 1.8 3.3 4.5 6.3 10
Solution:
Calculation of Trend by the method of least squares
Year y x x2 xy Trend value
Yc = a + bx

2000 1 –5 25 –5 0.223
2001 1.8 –3 9 –5.4 1.927
2002 3.3 –1 1 –3.3 3.631
2003 4.5 +1 1 4.5 5.335
2004 6.3 +3 9 18.9 7.037
2005 10 +5 25 50 8.743

y = 26.9x = 0 x2 = 70xy = 57.7

We know, trend value Yc = a + bx

y 26.9
a=   4.483
n 6

xy 59.7
b=   0.852
x 2 70

For the year Yc = a + bx

2000 = 4.483 + 0.852 (–5) = 0.223

Amity Directorate of Distance and Online Education

144 Quantitative Techniques in Management

Notes 2001 = 4.483 + 0.852 (–3) = 1.927

2002 = 4.483 + 0.852 (–1) = 3.631
2003 = 4.483 + 0.852 (+1) = 5.335
2004 = 4.483 + 0.852 (+3) = 7.037
2005 = 4.483 + 0.852 (+5) = 8.743
Illustration - 3
Below are given the annual production (in thousand tonnes) of a fertiliser factory:
Year Production
1992 70
1993 75
1994 90
1995 91
1996 95
1997 98
1998 100
(i) Fit a straight line trend by the method of least squares and tabulate the trend
values.
(ii) Estimate the likely production for the year 2000.
Solution:
Year y x (x2) (xy) Yc = a + bx
1992 70 –3 9 – 210 73.31
1993 75 –2 4 – 150 78.35
1994 90 –1 1 – 90 83.39
1995 91 0 0 0 88.43
1996 95 1 1 95 93.43
1997 98 2 4 196 98.51
1998 100 3 9 300 103.55
2
n=7  y = 619  x = 28  xy = 141

We know, Trend value, Yc = a + bx

y xy
a= and b =
n x 2

619 141
a= = 88.43 b= = 5.04
7 28
For the year Yc = a + bx
1992 = 88.43 + 5.04 (–3) = + 73.31
1993 = 88.43 + 5.04 (–2) = + 78.35

Amity Directorate of Distance and Online Education

Forecasting Techniques 145

1994 = 88.43 + 5.04 (–1) = + 83.39 Notes

1995 = 88.43 + 5.04 (0) = + 88.43
1996 = 88.43 + 5.04 (1) = 93.47
1997 = 88.43 + 5.04 (2) = + 98.51
1998 = 88.43 + 5.04 (3) = 103.55
(ii) The likely production for the year 2000
Y2000 = 88.43 + 5.04 (5) = 113.63 thousand tonnes
Illustration - 4
Find the following table:
(i) Fit a straight line trend by the method of least squares.
(ii) Calculate the trend values.

Year Production (in ‘000 tons)

1990 12
1991 10
1992 14
1993 11
1994 13
1995 15
1996 16
Solution:

Year y x (x2) (xy) Yc

1990 12 –3 9 –36 10.75

1991 10 –2 4 – 20 11.5
1992 14 –1 1 – 14 12.25
1993 11 0 0 0 13
1994 13 1 1 13 13.75
1995 15 2 4 30 14.5
1996 16 3 9 48 15.25
2
n=7  y = 91  x = 28  xy = 21

Straight line trend:

y xy
a= and b=
n x 2

91 21
a   13 b   0.75
7 28
Amity Directorate of Distance and Online Education
146 Quantitative Techniques in Management

Notes For the year Yc = a + bx

1990 = 13 + 0.75 (–3) = 10.75
1991 = 13 + 0.75 (–2) = 11.5
1992 = 13 + 0.75 (–1) = 12.25
1993 = 13 + 0.75 (0) = 13
1994 = 13 + 0.75 (1) = 13.75
1995 = 13 + 0.75 (2) = 14.5
1996 = 13 + 0.75 (3) = 15.25
Illustration - 5
The following figures represent the Annual sales of M/s. Suman Roy Company.
a) Find the trend values for each year, by adopting the method of
least squares.
b) Estimate the Annual sales for the next five years.

Year: 1996 1997 1998 1999 2000

Sales in lakhs: 46 50 40 70 60
Solution:

Year Sales Deviation Square of Derivation & Trend Value

y x deviation(x2) sales (xy) Yc

1996 46 –2 4 –92 43.6

1997 50 –1 1 –50 48.4
1998 40 0 0 0 53.2
1999 70 1 1 70 58
2000 60 2 4 120 62.8

2
n = 5  y = 266 x = 0 x = 10  xy = 48

y xy
Yc = a + bx a  n and b  2
x

266 48
a ; b  ; a = 53.2 and b = 4.8
5 10

For the year Yc = a + bx

1996 = 53.2 + 4.8 (–2) = 43.6
1997 = 53.2 + 4.8 (–1) = 48.4
1998 = 53.2 + 4.8 (0) = 53.2
1999 = 53.2 + 4.8 (1) = 58
2000 = 53.2 + 4.8 (2) = 62.8
2001 = 53.2 + 4.8 (3) = 67.6

Amity Directorate of Distance and Online Education

Forecasting Techniques 147

2002 = 53.2 + 4.8 (4) = 72.4 Notes

2003 = 53.2 + 4.8 (5) = 77.2
2004 = 53.2 + 4.8 (6) = 82
2005 = 53.2 + 4.8 (7) = 86.8
Illustration - 6
Calculate the trend values by applying least squares method.

Year: 1996 1997 1998 1999 2000 2001 2002

Sales: 20 23 22 25 26 29 30
(` Crores)

Estimate the likely sales for 2006.

Solution:

Year Sales x x2 xy Trend Value

(Yc = a+bx)
1996 20 -3 9 -60 20.08
1997 23 -2 4 -46 21.72
1998 22 -1 1 -22 23.36
1999 25 0 0 0 25
2000 26 1 1 26 26.64
2001 29 2 4 58 28.28
2002 30 3 9 90 29.92

2
n = 7  y = 175 x = 0  x = 28  xy = 46

We know, Trend Value, Yc = a+bx

y xy
a = b =
n x2

175 46
a = b =
7 28
= 25 = 1.64
For the year, Yc = a + bx
1996 = 25 + 1.64(-3) = 20.08
1997 = 25 + 1.64(-2) = 21.72
1998 = 25 + 1.64(-1) = 23.36
1999 = 25 + 1.64(0) = 25
2000 = 25 + 1.64(1) = 26.64
2001 = 25 + 1.64(2) = 28.28
2002 = 25 + 1.64(3) = 29.92
2006 is Yc = a + bx = 25 + 1.64(7) = 36.48

Amity Directorate of Distance and Online Education

148 Quantitative Techniques in Management

Notes I llustration - 7

Determine the trend values by the method of least squares from the following data Show the
actual time and trend line on graph.

Year: 1999 2000 2001 2002 2003 2004 2005

Values: 75 67 68 65 50 54 41
Solution:
(y) (x) (x2) Square of Trend value
Year Value Deviation deviation xy Yc

1999 75 -3 9 -225 75.63

2000 67 -2 4 -134 70.42
2001 68 -1 1 -68 65.21
2002 65 0 0 0 60
2003 50 1 1 50 54.79
2004 54 2 4 108 49.58
2005 41 3 9 123 44.37

2
N = 7  y = 420 x = 0  x = 28  xy = -146  Yc = 420

Where, Yc = a + bx
y 420 xy 146
a = = = 60 b = = = -5.21
N 7 x2 28

For the year, Yc = a + bx

1999 = 60 + (-5.21) (-3) = 75.63
2000 = 60 + (-5.21) (-2) = 70.42
2001 = 60 + (-5.21) (-1) = 65.21
2002 = 60 + (-5.21) (0) = 60
2003 = 60 + (-5.21) (+1) = 54.79
2004 = 60 + (-5.21) (2) = 49.58
2005 = 60 + (-5.21) (3) = 44.37

3.15 Summary
Correlation means that between two series or groups of data there exists some causal
connections.” At another place, he says, “If it is proved true that in a large number of instances
two variables tend always to fluctuate in the same or in opposite directions, we consider that the
fact is established and that a relationship exists. The relationship is called correlation.
Scatter Diagram is a graph of observed plotted points where each points represents the values
of X & Y as a co-ordinate. It portrays the relationship between these two variables graphically.
Probable error of Co-efficient of Correlation gives us the two limits within which the co-efficient
of correlation of Series selected at random from the same universe is likely to fall.

Amity Directorate of Distance and Online Education

Forecasting Techniques 149

Regression Analysis is mathematical measure of average relationship between two Notes

or more variables. Regression analysis is a statistical tool used in prediction of value of
unknown variable from known variable. Regression Analysis is a very powerful tool in the
field of statistical analysis in predicting the value of one variable, given the value of another
variable, when those variables are related to each other.
Time series analysis are basic to understanding past behaviour, evaluating current
accomplishments, planning for future operations and comparing different time series. Thus
a series of successive observations of the same phenomenon over a period of time are
called “time series”.
Seasonal variations in general refer to annual periodicity in business as economic
theory, the ideas involved can be extended to include periodicity over any interval of time
such as daily, hourly, weekly etc., depending on the type of data available.
Irregular or Random Analysis refer to the sporadic motions of time series due to
chance events such as floods, strikes, elections, fires, earthquakes, unusual weather etc.
Random variations in a time series are variations which cannot be accounted for by the
secular trend, seasonal movements or cyclical fluctuations.

3.16 Check Your Progress

I. Fill in the Blanks

1. _________ is the statistical method to determine the relationship between two
or more variables.
2. Variables whose measurement is done in terms such as weight, height and
length are classified as _________________
3. Minimum value of correlation is ______ .
4. In __________ correlation, the relation between more than 2 variables are
considered
5. Second differencing in time series can help to eliminate ___________ trend?
6. In time-series analysis, ___________ source of variation can be estimated by
the ratio-to-trend method?

II. True or False

1. Correlation is the statistical method to determine the relationship between two
or more variables.
2. I f increase in one variable leads to increase in dependent variable, then the
correlation is Positive.
3. Maximum value of correlation is -1.
4. In single correlation, the relation of two or more variables are considered
assuming other variables to be constant.
5. Spearman’s rank correlation method is based on Rank.

III. Multiple Choice Questions

1. A process by which we estimate the value of dependent variable on the basis
of one or more independent variables is called:
[a] Correlation
[b] Regression
Amity Directorate of Distance and Online Education
150 Quantitative Techniques in Management

Notes [c] Residual

[d] Slope
2. The method of least squares dictates that we choose a regression line where
the sum of the square of deviations of the points from the lie is:
[a] Maximum
[b] Minimum
[c] Zero
[d] Positive
3. A relationship where the flow of the data points is best represented by a curve
is called:
[a] Linear relationship
[b] Nonlinear relationship
[c] Linear positive
[d] Linear negative
4. The value we would predict for the dependent variable when the independent
variables are all equal to zero is called:
[a] Slope
[b] Sum of residual
[c] Intercept
[d] Difficult to tell
5. To determine the height of a person when his weight is given is:
[a] Correlation problem
[b] Association problem
[c] Regression problem
[d] Qualitative problem
6. When regression line passes through the origin, then:
[a] Intercept is zero
[b] Regression coefficient is zero
[c] Correlation is zero
[d] Association is zero
7. The predicted rate of response of the dependent variable to changes in the
independent variable is called:
[a] Slope
[b] Intercept
[c] Error
[d] Regression equation
8. Spearman’s rank correlation method is based on
[a] Values
[b] Samples
[c] Rank
[d] None of these

Amity Directorate of Distance and Online Education

Forecasting Techniques 151

3.17 Questions and Exercises Notes

I. Short Answer Questions

1. What is Correlation?
2. What is ‘Positive and ‘Negative Correlation’?
3. What is rank correlation?
4. What is a linear correlation?
5. What do you mean by coefficient of correlation?
6. What is a Probable Error?
7. Define the term regression.
8. What are the regression lines?
9. Define time series.
10. What is time series?
11. What do you mean by analysis of Time series?
12. What is Secular Trend?
13. What is seasonal variation?
14. What is Cyclical variation?
15. What Irregular variation?

II. Extended Answer Questions

1. Find the probable error when number of items are 5 and correlation is 0.7 for
a given distribution.
2. State the assumptions of Karl Pearson’s Coefficient Correlation.
3. Write down the Spearman’s Rank Coefficient Correlation Formula.
4. Define Rank Correlation Co-efficient.
5. Explain types of correlation.
6. Explain the properties of regression coefficients with examples.
7. Explain the components of time series.
8. Explain the merits of the least squares method.
9. What are the uses of analysis of time series?
10. Discuss the methods of measuring trend values.

3.18 Key Terms

 Correlation analysis: The statistical technique which deals with the
association between two or more variables is known as correlation analysis.
 Regression: Regression is the measure of the average relationship between
two or more variables in terms of the original units of the data.
 Regression analysis: The term regression analysis refers to the methods by
which estimates are made of the values of a variable from a knowledge of the
values of one or more other variables and to the measurement of the errors
involved in this estimation process.
 Time series: A time series consists of statistical data which are collected,
recorded or observed over successive increments.

Amity Directorate of Distance and Online Education

152 Quantitative Techniques in Management

Notes  Cyclical Analysis: Its refer to the long term swing about a trend line or curve.
These cycles, as they are sometimes called may or may not be periodic, i.e.,
they may or may not follow exactly similar pattens after equal intervals of time.
 Random Analysis: Random or erratic variations are variations which occur in
a completely unpredictable fashion.

3.19 Check Your Progress: Answers

I. Fill in the Blanks
1. Correlation
2. Continuous variables
3. Minus one
4. Multiple
5. Quadratic Trend
6. Seasonal
II. True or False
1. True
2. True
3. False
4. False
5. True
III. Multiple Choice Questions
1. (b)
2. (b)
3. (a)
4. (c)
5. (c)
6. (a)
7. (a)
8. (c)

3.20 Case Study

For the following table,
a) Fit a straight line trend by the method of least squares
b) Show the trend line on a graph
c) Estimate the income for the year 2010
Year: 2000 2001 2002 2003 2004 2005 2006
Income: 300 700 600 800 900 700 1000
(` 000)

Amity Directorate of Distance and Online Education

Forecasting Techniques 153

3.21 Further Readings Notes

1. Richard I. Levin, David S. Rubin, Sanjay Rastogi Masood Husain Siddiqui,
Statistics for Management, Pearson Education, 7th Edition, 2016.
2. Prem.S.Mann, Introductory Statistics, 7th Edition, Wiley India, 2016.
3. Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani, An Introduction
to Statistical Learning with Applications in R, Springer, 2016.
4. Aczel A.D. and Sounderpandian J., “Complete Business Statistics”, 6th edition,
Tata McGraw Hill Publishing Company Ltd., New Delhi, 2012.
5. Anderson D.R., Sweeney D.J. and Williams T.A., Statistics for business and
economics, 11th edition, Thomson (South – Western) Asia, Singapore, 2012.

3.22 Bibliography
1. Srivastava V. K. etal – Quantitative Techniques for Managerial Decision Making,
Wiley Eastern Ltd
2. Richard, I.Levin and Charles A.Kirkpatrick – Quantitative Approaches to
Management, McGraw Hill, Kogakusha Ltd.
3. Prem.S.Mann, Introductory Statistics, 7th Edition, Wiley India, 2016.
4. Aczel A.D. and Sounderpandian J., “Complete Business Statistics”, 6th edition,
Tata McGraw Hill Publishing Company Ltd., New Delhi, 2012.
5. Anderson D.R., Sweeney D.J. and Williams T.A., Statistics for business and
economics, 11th edition, Thomson (South – Western) Asia, Singapore, 2012.
6. Budnik, Frank S Dennis Mcleaavey, Richard Mojena – Principles of Operation
Research - AIT BS New Delhi.
7. Sharma J K – Operation Research- theory and applications-Mc Millan,New Delhi
8. Kalavathy S. – Operation Research – Vikas Pub Co
9. Gould F J – Introduction to Management Science – Englewood Cliffs N J Prentice
Hall.
10. Naray J K, Operation Research, theory and applications – Mc Millan, New Dehi.
11. Taha Hamdy, Operations Research, Prentice Hall of India
12. Tulasian: Quantitative Techniques: Pearson Ed.
13. Vohr.N.D. Quantitative Techniques in Management, TMH.
11. Stevenson W.D, Introduction to Management Science, TMH.


Amity Directorate of Distance and Online Education

154 Quantitative Techniques in Management

Notes

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 155

Notes

Unit 4: Probability and Testing of Hypothesis

Structure:
4.1 Introduction of Probability Theories
4.2 Concepts
4.3 Addition and Multiplication Theorems
4.4 Probability Distributions
4.5 Sampling
4.6 Steps in Developing a Sampling Plan
4.7 Sampling Techniques
4.8 Merits and Limitations of Sampling
4.9 Sampling Distribution
4.10 Hypothesis
4.11 Null Hypothesis and Alternative Hypothesis
4.12 t – test
4.13 F – test
4.14 Non–Parametric Tests (chi – square test)
4.15 Summary
4.16 Check Your Progress
4.17 Questions and Exercises
4.18 Key Terms
4.19 Check Your Progress: Answers
4.20 Case Study
4.21 Further Readings
4.22 Bibliography

Objectives

After studying this unit, you should be able to understand:

 Concept of Probability Theories
 Concepts, Addition and Multiplication Theorems
 Overview of Probability Distributions
 Concept of Sampling and Sampling Distribution
 Concept of Null Hypothesis and Alternative Hypothesis,
 Overview of t – test and F – test
 Concept of Non – Parametric Tests (chi – square test)

Amity Directorate of Distance and Online Education

156 Quantitative Techniques in Management

Notes 4.1 Introduction of Probability Theories

Every human activity involves certain element of uncertainty. Uncertainty like probably
it may rain today, probably inflation rate will be high in coming years or probably it may
get number one rank in class. Therefore, there is a need to handle uncertainty
systematically and scientifically.
a) Probability is a numerical measure of the likelihood that an event will occur.
b) Probability measures the degree of uncertainty.
c) Probability values are assigned on a scale of 0 to 1.

Definition of Probability

Probability of a given event is an expression of likelihood or chance of occurrence

of an event. A probability is a number which rages from zero to one.
Probability is a numerical measure which indicates the chance of occurrence of an
event ‘A’. It is denoted by P(A). It is the ratio between the favorable outcomes of an event
‘A’ (m) to the total outcomes of the experiment (n).

m
P(A) =
n

Where, ‘m’ is the number of favourable outcomes of an event ‘A’ and ‘n’ is the total
number of outcomes of the experiment.

Meaning of Probability

Probability is the likelihood that an event will occur and is calculated by dividing the
number of favourable outcomes by the total number of possible outcomes.
Example: There are only two possible outcomes when a coin is flipped, the result
is either heads or tails. And so the probability of getting heads is 1 out of 2 or ½ (50%).

4.2 Concepts of Probability

Experiment:

Experiment refers to describe an act which can be repeated under some given
conditions. Random experiments are those experiments whose results depend on chance
such as tossing a coin, throwing a dice etc. The outcomes of random experiment are
events.
As previously stated, an experiment is a process that produces outcomes. Examples
of business oriented experiments with outcomes that can be statistically analyzed might
include the following:
1. Interviewing 20 randomly selected consumers and asking them which brand of
appliance they prefer.
2. Sampling every 200th bottle of ketchup from an assembly line and weighing
the contents.
3. Testing new pharmaceutical drugs on samples of cancer patients and measuring
the patient's improvement.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 157

4. Auditing every 10th account to detect any errors. Notes

5. Recording the Industrial Average on the first Monday of every month for 10 years.

Random Experiment:

When the outcome of an experiment cannot be predicted with certainty, then it is

called random experiment or stochastic experiment.

Sample Space:

The set of all possible outcomes of a random experiment is the sample space. The
sample space is denoted by S. The outcome of the random experiment are sample points
or outcomes or cases. For examples: In tossing of two coins the outcome are Head &
Tail
Hence Sample Space ‘S’ = {HH, HT, TT, TH}

Events:

An event is a possible outcome of an experiment or a result of trial or an observation.

Basic two types of events are. Sure and Impossible event and Random event. An event
whose occurrence is inevitable when a certain random. Experiment is performed is called
a certain or sure event. An event which can never occur when a certain random experiment
is performed is called an impossible event. An event which may or may not occur while
performing a certain random experiment is known as a random event.
It is a subset of the Sample Space. Events are denoted by A, B, C for example
when throwing a die, A = {1, 3 ,5} is an event.
Probability of occurrence of an event:
Let, S be the sample and let E be an event.

Mutually exclusive Events:

Two events are said to be mutually exclusive or incompatible when both cannot
happen simultaneously in a single trial. i.e., the occurrence of any one of them precludes
the occurrence of the other. For example, if a single coin is tossed, either head or tail
can turn up. If we consider at one point of time, a person can be alive or dead.

Independent and Dependent Events:

Two or more events are said to be independent when the outcome of one does not
affect, and is not affected by the other. For example, if a coin is tossed twice, the result
of second would not be affected by the result of the first tossing.
Dependent events are those in which the occurrence or non-occurrence of one events
in other trails. For example, if a card is drawn from a pack of playing cards and is not
replaced, this will alter the probability that the second card is drawn.

Equally likely Events:

Events are said to be equally likely when one does not occur more often than the
others. For example, if an unbiased dice is thrown, each face may be expected to be
observed approximately the same number of times in the long run.

Amity Directorate of Distance and Online Education

158 Quantitative Techniques in Management

Notes Simple and Compound Events:

In case of simple events, consider the probability of the happening or not happening
of single events. For example: If the are choosing two black coins from a box containing
10 white and 5 black coins.
In case of compound events, consider the joint occurrence of two or more events.
For example, if are interested in choosing or drawing 3 white coins in first draw and 2
black coins in second draw from a bag containing 10 white and 5 black coins.

Exhaustive Events:

Events are said to be exhaustive when their totality includes all the possible
outcomes of a random experiment. For example, while throwing a dice, the possible
outcomes are 1, 2, 3, 4, 5 and 6 and hence the exhaustive number of cases is 6.

Complementary Events:

If A is an event of the number of favourable cases in the experiment, then A is

called the complementary event of non-favourable cases in the experiment. Thus, the
events A and A are mutually exclusive and collectively exhaustive events.

Elementary Events:

Events that cannot be decomposed or broken down into other events are called
elementary events. Elementary events are denoted by lowercase letter.

Combinations of Events

The complement Ac of an event A is the event that A does not occur

Probability Rule:
P(Ac) = 1 - P(A)
The union of two events A and B is the event that either A or B or both occurs
The intersection of two events A and B is the event that both A and B occur

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 159

Notes

Disjoint Events

Two events are called disjoint if they can not happen at the same time
Events A and B are disjoint means that the intersection of A and B is zero
Example: Coin is tossed twice
S = {HH, TH, HT, TT}
Events A = {HH} and B = {TT} are disjoint
Events A = {HH, HT} and B = {HH} are not disjoint
Probability Rule: If A and B are disjoint events then
P(A or B) = P(A) + P(B)

4.3 Addition and Multiplication Theorems

Additional Theorem

The addition theorem of probability states that if two events A and B are mutually
exclusive, the probability of occurrence of either A or B is sum of the individual probability
of A and B.
The rule of addition applies to the following situation. When there are two events,
the probability that either event occurs is to be known.
If A and B are two events associated with an experiment, then probability of
occurence of events A or B or both A and B is given by
P(A  B) = P(A) + P(B) – P(A  B)

Corollary 1:

If A and B are mutually exclusive events, then A  B =  . Hence,

n ( A  B) 0
P(A  B) = n ( S) = n( S ) = 0

 Addition theorem, in this case, will be P(A  B) = P(A) + P(B).

Corollary 2:

If A and A are complementary events, then A  A = S and A  A = 

Amity Directorate of Distance and Online Education

160 Quantitative Techniques in Management

Notes  P(A  A ) = P(S) . . . (i)

n( S )
But, P(S) =
n( S ) = 1 . . . (ii)

and by addition theorem,

P(A  A ) = P(A) + P( A ) . . . (iii) [  A  A =  ]
 from (1), (2) and (3),
P(A) + P( A ) = 1

 P( A ) = 1  P(A)

Corollary 3:

Addition theorem on probability, given above, can be extended to there events A,

B and C. Probability of occurance of atleast one of the three events A, B and C is given
by
P(A  B  C) = P(A) + P(B) + P(C)  P(A  B)  P(B  C)
 P(A  C) + P(A  B  C)
Note: If events A, B and C are mutually exclusive events, then
P(A  B  C) = P(A) + P(B) + P(C)
In general, if A1, A2, . . . . An are n mutually exclusive events associated an
experiment, the probability of occurence of atleast one of these n events is given by
P(A1  A2  . . . .  An) = P(A1) + P(A2) + . . . + P(An)

Corollary 4:

From the following diagram it will be clear that A  B and b A  Bg are

complementary events of each other.

 (A  B)  bA  Bg = S and (A  B)  bA  Bg = 
 By Corollary 2,

b g
P A  B = 1  P(A  B) i.e., P(A  B) = 1  P A  B b g
This Corollary is very important, as it is useful in solving many examples. It can be
interpreted as
P(Occurence of atleast one event) 1  P (Occurence of none of the events)

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 161

Independent Events Notes

Events A and B are independent if probability of event A is not affected by occurrence
or non-occurrence of event B.
In such a case P(A/B) = P(A) and P(B/A) = P(B)
Example:
Tossing two coins, Event A = first coin is a head. Event B = second coin is a head
Disjoint events cannot be independent! If A and B cannot occur together (disjoint),
then knowing that A occurs does change probability that B occurs.

Example:

If A and are independent events then prove that A and B are independent events.

Solution:

Given, A and B are independent

i.e., P(A  B) = P(A), P(B)
TPT P(A  B ) is dependent

Consider P(A  B ) = P(A) – P(A  B)

= P(A) – P(A) P(B)
= P(A) [1 – P(B)]
= P(A), P( B )
 Independent

Multiplication Theorem of Probability

The rule of multiplication applies to the situation when the probability of the
intersection of two events is to be known that is, the probability that two events (Event
A and Event B) both occur.
If A and B are any two events associated with an experiment, then the probability
of simultaneous occurence of events A and B is given by
P(A  B) = P(A) . P(B/A) = P(B) . P(A/B)

Corollary 1:

If A and B are independent events, then P(A/B) = P(A) and P(B/A) = P(B), hence
multiplication theorem in such a case will be
P(A  B) = P(A) . P(B)
Note: If A, B and C are there independent events associated with an experiment,
then
P(A  B  C) = P(A) . P(B) . P(C)
In general, if A1, A2, . . . . An are n independent events, then probability of
simultaneous occurence of these n events is
P(A1  A2  . . . .  An) = P(A1) P(A2) . . . P(An)

Amity Directorate of Distance and Online Education

162 Quantitative Techniques in Management

Notes Corollary 2:

If A and B are mutually exclusive events, P(A  B) = 0

Also, if A occurs B cannot occur.
 P(B/A) = 0
Similarly, P(A/B) = 0
Odds in Favour and Against:
If the probability of occurence of an event A is p then the probability of non-occurence
of A is (1  p). The odds in favour of A is given by p : (1  p) : p.
Example:

2 2 3
If P(A) = , then P( A ) = 1  P(A) = 1  =
5 5 5
The odds in favour of A is given by

2 3
: i.e., 2 : 3
5 5
The odds against A is given by

3 2
: i.e., 3 : 2
5 5
Conversely, if odds in favour of A are a : b, then to find the probability of A, we divide
by (a + b).

a b
i.e., odds in favour of A can be written as
ab : ab

a b
 P(A) = a  b and P( A ) = a  b

Practical Problems on Probability

Illustration - 1

A ball is drawn at random from a box containing 6 red balls, 4 white balls and 5
blue balls. Find the probability that the ball drawn is not red.

Solution:

Total number of balls = 6 + 4 + 5 = 15

6
Probability of drawing a red ball =
15

6 9
Probability of not red = 1  =
15 15

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 163

Illustration - 2 Notes
In a single throw of two dice what is the probability of getting a total of 9?
Solution:
When two dice are thrown, the possible combinations are 62 = 36,
The cases which have a total of 9 are (3, 6), (4, 5), (5, 4), (6, 3) = 4

4 1
 The probability = 36  9

Illustration - 3

A pair of unbiased dice is thrown. If the two numbers appearing be different, find
the probabilities that (i) the sum 6 (ii) the sum is 5 or less.
Solution:

(i) The Sum:

If two dice are thrown there are 36 (= 6 × 6) exhaustive, mutually and equally likely
outcomes (i) favourable cases are (1, 5), (2, 4), (4, 2), (5, 1) i.e. only 4 cases.

4 1
P(total six in sum) = =
36 9

(ii) Favorable case are:

For sum 5 : (1, 4), (2, 3), (3, 2), (4, 1)

For sum 4 : (1, 3). (3, 1)
For sum 3 : (1, 2), (2, 1) Find 8 favourable cases

8 2
 Required probability = =
36 9

Illustration - 4

If 8 persons are arranged at random (i), in a line, (ii) in a ring find the probability
that two particular person will be always together.

Solution:

(i) 8 persons can be arranged in a line 8! ways, which are mutually exclusive,
exhaustive and equally likely.
Now taking 2 particular person as one, 7 = (1 + 6) person can be arranged in 7!
Ways. Again these persons can be arranged almost themselves in 2! Ways. So, the total
number of favorable cases 7! 2!

7! 2! 1
 Required probability = =
8! 4

Amity Directorate of Distance and Online Education

164 Quantitative Techniques in Management

Notes (ii) 8 persons can be arranged in a ring in 7! Ways. As before number of favourable
cases is 6! 2!

6! 2! 2
 Required probability = =
7! 7

Illustration - 5

A company wants to select a team leader from among the following candidates:
a) Male, age 40
b) Male, age 43
c) Female, age 38
d) Female, age 29
e) Male, age 39
What is the probability that the team leader selected, will be
1) Either male or aged above 41 years.
2) Either female or aged below 35 years.

Solution:

(1) P(male) above 41 years

= P(male) + P(above 41) – (male and above 41)

3 1 1 4 1 3
=   =  =
5 5 5 5 5 5
(2) P(female) below 35 years

2 1 1 3 1 2
=     =
5 5 5 5 5 5

Illustration - 6

What is probability of getting 3 white balls in a draw of 3 balls from a box containing
5 white and 4 black balls?

Solution:

Total number of balls in the box 5 + 4 = 9.

987
3 balls can be drawn from 9 balls in 9C3 =
3  2  1 = 84 ways
 Total number of possible cases = 84.
Total number of white balls in the box = 5.

543
3 white balls can be drawn from 5 white balls in 5C3 =
3  2  1 = 10 ways

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 165

Number of cases favourable to the event of getting 3 white balls = 10. Notes
10 5
Probability of drawing 3 white balls = =
84 42

Illustration - 7

5 men in a batch of 20 are graduates. If 3 men are picked up out of 20 at random,

what is the probability that they are all graduates? What is the probability of at least one
graduate?

Solution:

1st part: 3 men can be selected out of 20 men in 20C3 ways, so total number of
possible cases is 20C3. Again 3 men to be graduates, these 3 men are to be selected
out of 5 graduates, which can be done in 5C3. Now the total number of favourable cases
is 5C3.
5
C3 10 1
Required probability = 20 = =
C3 1140 114

5! 3! 5. 4 .3
Calculation: 5C3 = =
2! 3. 2 = 10

20.19.18
20C = = 1140
3 3. 2
2nd Part: At least one graduate means it may be 1, 2 or 3 graduates. now it will
be easier to find the complementary event i.e. none is graduate (or all 3 are non-graduate).
So 3 men (non-graduate) can be selected out 15 ( = 10 – 5) non-graduates in 15C3 ways.
Total number of favourable cases is 15C3.
15
C3 455 91 137
Required probability = 1  20 =1  =1  =
C3 1140 228 228

15.14 .13
Calculation: 15C = = 455,
3 3. 2.1

20.19.18
15C =
3 3. 2.1 = 1140

Illustration - 8

An urn contains 8 red and 3 blue marbles. If 2 marbles are drawn at random, find
the probability that (i) Both are red and (ii) One of them is red and another is blue.
Solution:
Total number of marbles in the urn = 8 + 3 = 11
11  10
11
Two marbles can be drawn from 11 in C2 ways =
2  1 = 55 ways

Amity Directorate of Distance and Online Education

166 Quantitative Techniques in Management

Notes (i) Probability that both the marbles are red:

87
2 marbles can be drawn from 8 red marbles in 8C2 ways =
2  1 = 28 ways

28
The probability that the two marbles drawn at random are red = = 0.509
55
(ii) Probability that one of them is red and another is blue:
One red marble can be drawn from 8 red marbles in 8c1 ways and 1 blue marble can
be drawn from 3 blue marbles in 3c1 ways.
The number of favourable cases is 8c1 x 3c1 and the required probability is

8c1 x 3c1 8 x3 24
P = = = = 0.436
11c 2 55 55

Illustration - 9

Three cards are drawn at random from a pack of 52 cards. Find the probability that:
(i) They are king, Queen and Knave
(ii) All of them are Aces
(iii) All of them are Spades
(iv) Two are red and one Black
(v) They are not face cards.

Solution:

There cards can be drawn at random from a well shuffled pack of 52 cards in 52C3
ways.
(i) One King can be drawn out of 4 Kings (in a full pack) in 4C3 ways. Similarly one
Queen in 4C3 and one Knave in 4C3 ways. Since each way of drawing a King can be
associate with each of way of drawing Queen and Knave, so number of favourable cases
is 4C1 × 4C1 × 4C1
4
C1  4 C1  4 C1 444 16
Required probability = 52 = =
C3 22100 5525
(ii) There Aces can be drawn out of four Aces in 4C3 ways
4
C3
Required probability = 52
C3
(iii) 3 Spades can be drawn out of 13 Spades in 13C3 ways.
13
C3
Required probability = 52
C3

26 26
C2  C1
(iv) Required probability = 52
C3

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 167

(v) Number of cards other than face = 52 – 12 = 40, as there are 12 face cards. Notes
40
C3
Required probability = 52
C3

Illustration - 10

What is the probability of getting a Jack or a Spade from a pack of 52 cards.

Solution:

There are 4 Jacks and 13 Spades in a pack of 52 cards. Also Spade has one Jack in
it.

13
 Probability of getting a Spade =
52

4
 Probability of getting a Jack =
52

1
 Probability of getting a Jack Spade =
52
Let, A be the event of getting a Spade
B be the event of getting a Jack.
As there is a Jack in Spade, the events are not mutually exclusive events. Hence,
the events are not mutually exclusive events. Hence, the probability of getting a Jack or a
Spade is
P(A  B) = P(A) + P(B) – P(A  B)

13 4 1 16 4
=    
52 52 52 52 13

4.4 Probability Distributions

A probability distribution is a statistical function. It describes all the possible values
and possibilities that a random variable can take within a given range. This range will be
in between the minimum and maximum statistically possible values. But, it is where the
possible value is likely to be plotted on the probability distribution depends on a number
of factors. These factors include the distribution's mean, standard deviation, skewness
and kurtosis.
Probability Distribution maps out the likelihood of multiple outcomes in a table or
an equation. In the previous example of flipping a coin which had the probability outcome
of head or tail. But, if the coin is flipped twice in a row, there are four possible outcomes
(heads-heads, heads-tails, tails-heads and tails-tails). So there are a series of potential
outcomes.

Amity Directorate of Distance and Online Education

168 Quantitative Techniques in Management

Notes Definitions
1. Two events are mutually exclusive or disjoint if they cannot occur at the same
time.
2. The probability that Event A occurs, given that Event B has occurred, is called
a conditional probability. The conditional probability of Event A, given Event B,
is denoted by the symbol P (A|B).
3. The complement of an event is the event not occurring. The probability that Event
A will not occur is denoted by P (A').
4. The probability that Events A and B both occur is the probability of the
intersection of A and B. The probability of the intersection of Events A and B
is denoted by P (A  B). If Events A and B are mutually exclusive, P (A  B)
= 0.
5. The probability that Events A or B occur is the probability of the union of A
and B. The probability of the union of Events A and B is denoted by P(A 
B).
6. If the occurrence of Event A changes the probability of Event B, then Events
A and B are dependent. On the other hand, if the occurrence of Event A does
not change the probability of Event B, then Events A and B are independent.

Discrete and Continuous Distribution

All probability distributions can be classified as discrete probability distributions or

as continuous probability distributions, depending on whether they define probabilities
associated with discrete variables or continuous variables.
Discrete Probability Distributions
The discrete random variables are those that take on integer values only. They never
include fractions or decimals, they are separate.
If a random variable is a discrete variable, its probability distribution is called a
discrete probability distribution.
A discrete random variable is a random variable that has values that has either a
finite number of possible values or a countable number of possible values. It is usually
the result of counting something.
An example will make this clear. Suppose you flip a coin two times. This simple
statistical experiment have four possible outcomes: HH, HT, TH, and TT. Now, let the
random variable X represent the number of Heads that result from this experiment. The
random variable X can only take on the values 0, 1 or 2, so it is a discrete random variable.
Also give examples such as:
1. The number of students in a class.
2. Number of home mortgages approved by Coastal Federal Bank last week.
3. The number of children in a family.
4. The number of cars entering a car wash in a hour.
The following are discrete probability distributions:
1. Binomial probability distribution.
2. Hypergeometric probability distribution.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 169

3. Multinomial probability distribution. Notes

4. Negative binomial distribution.
5. Poisson probability distribution.

Continuous Probability Distributions

Continuous random variables are those that take on any value including fractions
and decimals. Continuous random variables give rise to continuous probability
distributions. Continuous is the opposite of discrete.
If a random variable is a continuous variable, its probability distribution is called a
continuous probability distribution.
A continuous random variable is a random variable that has an infinite number of
possible values that is not countable .It is usually the result of some type of measurement.
Example:
1. The income in a year for a family.
2. The distance students travel to class.
3. The time it takes an executive to drive to work.
4. The length of an afternoon nap.
5. The length of time of a particular phone call.
A continuous probability distribution differs from a discrete probability distribution in
several ways.
1. The probability that a continuous random variable will assume a particular value
is zero.
2. As a result, a continuous probability distribution cannot be expressed in tabular
form.
3. Instead, an equation or formula is used to describe a continuous probability
distribution.
Most often, the equation used to describe a continuous probability distribution is
called a probability density function. Sometimes, it is referred to as a density function,
a PDF or a pdf. For a continuous probability distribution, the density function has the
following properties:
(i) Since the continuous random variable is defined over a continuous range of
values i.e., called the domain of the variable, the graph of the density function
will also be continuous over that range.
(ii) The area bounded by the curve of the density function and the x-axis is equal
to 1, when computed over the domain of the variable.
(iii) The probability that a random variable assumes a value between a and b is equal
to the area under the density function bounded by a and b.

Expected Value and Variance

All probability distributions are characterized by an expected value (mean) and a

variance (standard deviation squared).
The expected value (or mean) of X, where X is a discrete random variable, is a
weighted average of the possible values that X can take, each value being weighted

Amity Directorate of Distance and Online Education

170 Quantitative Techniques in Management

Notes according to the probability of that event occurring. The expected value of X is usually
written as E(X) or m.
E(X) = S  P(X = x)
The expected value is the sum of: [(each of the possible outcomes) × (the probability
of the outcome occurring)].
In more concrete terms, the expectation is what you would expect the outcome of
an experiment to be on average.
Example:
What is the expected value when we roll a fair die?
There are six possible outcomes: 1, 2, 3, 4, 5, 6. Each of these has a probability
of 1/6 of occurring. Let X represent the outcome of the experiment.
Therefore P(X = 1) = 1/6 (this means that the probability that the outcome of the
experiment is 1 is 1/6)
P(X = 2) = 1/6 (the probability that throw a 2 is 1/6)
P(X = 3) = 1/6 (the probability that throw a 3 is 1/6)
P(X = 4) = 1/6 (the probability that throw a 4 is 1/6)
P(X = 5) = 1/6 (the probability that throw a 5 is 1/6)
P(X = 6) = 1/6 (the probability that throw a 6 is 1/6)
E(X) = 1 × P(X = 1) + 2 × P(X = 2) + 3 × P(X = 3) + 4 × P(X = 4) + 5 × P(X =
5) + 6 × P(X = 6)
Therefore, E(X) = 1/6 + 2/6 + 3/6 + 4/6 + 5/6 + 6/6 = 7/2
The Mean of a Probability Distribution
The mean is a typical value used to represent the central location of a probability
distribution.
The mean of a probability distribution is also referred to as its expected value.
The Variance and Standard Deviation of a Probability Distribution
The computational steps are:
1. Subtract the mean from each value, and square this difference.
2. Multiply each squared difference by its probability.
3. Sum the resulting products to arrive at the variance.

Binomial Distribution

If a group of patients is given a new drug for the relief of a particular condition, then
the proportion p being successively treated can be regarded as estimating the population
treatment success rate  .
The sample proportion p is analogous to the sample mean x , in that if we score
zero for those s patients who fail on treatment and unity for those r who succeed, then
p = r/n, where n = z + s is the total number of patients treated. Thus p also represents
a mean.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 171

Data which can take only a 0 or 1 response, such as treatment failure or treatment Notes
success, follow the binomial distribution provided the underlying population response rate
does not change. The binomial probabilities are calculated from
Probability of obtaining exactly ‘r’ successes in a given number of ‘n’ Bernoulli trials
is
f(r) = P(X = r) = n Cr pr qn - 1
For r = 1, 2, ….. n
Where,
p = probability of success on a single trial.
q = probability of failure on a single trial.
e = number of (Bernoulli) trials.
r = number of successes in ‘n’ trials.
Mean Binomial distribution (  ) = np
Variance of Binomial distribution(  2) = npq
For successive values of R from 0 through to n. In the above n! is read as n factorial
and R! as R factorial. For R = 4, R! = 4 × 3 × 2 × 1 = 24. Both 0! and 1! are taken
as equal to unity. The shaded area marked in Figure corresponds to the above expression
for the binomial distribution calculated for each of R = 8, 9, ..., 20 and then added. This
area totals 0.1018. So the probability of eight or more responses out of 20 is 0.1018.
For a fixed sample size n the shape of the binomial distribution depends only on
 . Suppose n = 20 patients are to be treated and it is known that on average a quarter
 or = 0.25 will respond o this particular treatment. The number of responses actually
observed can only take integer values between 0 (no responses) and 20 (all respond).
The binomial distribution for this case is illustrated in below figure.
The distribution is not symmetric, it has a maximum at five responses and the height
of the blocks corresponds to the probability of obtaining the particular number of responses
from the 20 patients yet to be treated. It should be noted that the expected value for r,
the number of successes yet to be observed if we treated n patients, is n  . The potential
variation about this expectation is expressed by the corresponding standard deviation
SE(r) = [n  (1-  )]
Figure shows the Normal distribution arranged to have µ = n  = 5 and s = v [n  (1-
 )] = 1.94, superimposed on to a binomial distribution with p = 0.25 and n = 20. The
Normal distribution describes fairly precisely the binomial distribution in this case.
If n is small, however or  close to 0 or 1, the disparity between the Normal and
binomial distributions with the same mean and standard deviation, similar to those
illustrated in Figure, increases and the normal distribution can no longer be used to
approximate the binomial distribution. In such cases the probabilities generated by the
binomial distribution itself must be used.
It is also only in situations in which reasonable agreement exists between the
distributions use the confidence interval expression given previously. For technical
reasons, the expression given for a confidence interval for  is an approximation. The
approximation will usually be quite good provided p is not too close to 0 or 1, situations
in which either almost none or nearly all of the patients respond to treatment. The
approximation improves with increasing sample size n.
Amity Directorate of Distance and Online Education
172 Quantitative Techniques in Management

Notes

Fig: Binomial distribution for n = 20 with = 0.25 and the Normal approximation

Illustration - 1

A merchant’s file of 20 accounts contains 6 delinquent and 14 non-delinquent

accounts. An auditor randomly selects 5 of these accounts for examination.
(i) What is the probability that the auditor finds exactly 3 delinquent accounts.
(ii) Find the expected number of delinquent accounts in the sample selected.
Solution:

6
Probability of a delinquent account P = = 0.3
20
Probability of a non-delinquent account q = 1 – p = 1 – 0.3 = 0.7
(i) Probability that auditor finds (r = 2) delinquent accounts in a random choice of(n = 5)
accounts is given by Binomial distribution as:
p(r) = n Cr pr qn - 1

= 5C2 (0.3)2 (0.7)5 - 3 = 5C2 (0.3)2 (0.7)2

= 10 × 0.09 × 0.49 = 0.441
(ii) Expected number of delinquent accounts in the selected sample of an n = 5 is given
by Mean = np = 5 × 0.3 = 1.5

Illustration - 2

The average percentage of failures in a certain examination is 40. What is the

probability that out of a group of 6 candidates, at least 4 passed in the examination?
What is the probability that at the most 2 passed?

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 173

Solution: Notes
Probability of success p = 0.60
(  P = 1 – q and q = 0.4)
Probability of failure q = 0.40 (given)
Number of trails = 6
By Binomial distribution, the probability that ‘r’ candidates the examination is

p(r) = n Cr pr qn - 1

= 6 n Cr pr qn - 1

(i) Probability that at least 4 passed is

P(X  4) = P(4) + P(5) + P(6)

= (6 C4 (0.6)4 (0.4)2) + (6 C5 (0.6)5 (0.4)1) + (6 C6 (0.6)6 (0.4)0)

= (15 × 0.1296 × 0.16) + (6 × 0.07776 × 0.4) + (0.04666)
= 0.3110 + 0.1866 + 0.0467 = 0.5443
ii) Probability that at the most 2 passed is
P(X  2) = P(0) + P(1) + P(2)

= (6C0 (0.6)0 (0.4)6) + (6C1 (0.6)1 (0.4)5) + (6C2 (0.6)2 (0.4)4)

= 0.0041 + 0.0369 + 0.1382

= 00.1792

Illustration - 3

The probability that an entering student will graduate is 0.4. Determine the probability
that out of 5 students at least one will graduate.

Solution:

P = 0.4 q = 0.6 n=5

On Binomial Distribution the probability of at least one will graduate is
P(X  1) = 1 – P(X < 1)
= 1 – P(X = 0)
= 1 – 5C0 (0.4)0 (0.6)5 = 1 – (0.6)5

Illustration - 4

Find the mean and variance of Binomial distribution with parameters n and p.

Solution:

Mean = E(X), Variance = E(X2) – [E(X)]2

Probability mass function is

Amity Directorate of Distance and Online Education

174 Quantitative Techniques in Management

Notes P(X = 2) = P(x) = nCx px qn

 x
, x = 0, 1, 2 …. n

n n

Mean = E(X) =  xp(x) =  x. nCx px qn

 x
x0 x0

=  n. nCx . px qn . q
x
x0


n
FG pIJ x

= qn
x0
x. nCx
H qK
L F pI F pI F pI F pI OP
= M 0 + 1 . nC G J + 2 . nC G J + 3 . nC G J +..+ n.nC G J
2 3 n

MN
qn
HqK
1
H qK 2 H q K HqK 3 n
PQ
LMn FG pIJ  2. n ( n 1) FG pIJ  3. n ( n  1) ( n  2) FG pIJ  ...  nFG pIJ OP
2 3 n

=
N HqK 2 H qK 3! H qK H q K PQ
F pI L
= q . n G J M 1 + (n  1) G J +
F pI ( n  1) ( n  2) FG pIJ + ... + FG pIJ 2 n 1
OP
n
H q K MN HqK 2! H qK HqK PQ
F pI L p O = q np LM( p  q ) OP = q . np(1)
= q . n G J M1  P
n 1 n 1 n
n
HqK N q Q q N q
n
Q q .q n 1 n 1

= np
n

E(X2) =  x2p(x)
x0

=  (x2  x + x) p(x)
x0

n n

=  (x2  x) p(x) +  xp(x)

x0 x0

n n

Now, =  (x2  x) p(x) =  x(x  1) p(x)

x0 x0

= 
 x
x(x  1) nCx px qn
x0


n
FG pIJ x

=
x0
x(x  1) nCx
H qK qn

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 175

LM F pI 2
FG pIJ 3
FG pIJ 4
FG pIJ n OP Notes
0 + 0 + 2.1nC G J
= qn
MN H qK 2 + 3.2nC3
H qK + 4.3nC4
H qK + ... + nCn
HqK PQ
LM FG IJ + 3.2 . ( n  1) ( n  2) FG pIJ + ... + n(n 1) FG pIJ
n ( n 1) p
2 3 n OP
= qn
MN 2.1
2 H K
q 3.21. H qK  HqK PQ
F pI L F pI ( n  2) ( n  3) FG pIJ + .. + FG pIJ
. n(n - 1) G J M 1 + (n - 2) G J +
2 2 n2 OP
= qn
H q K MN HqK 2 H qK H qK PQ
F pI L p O = q . n(n 1) p . ( p  q )
. n(n  1) G J M1  P
2 n2 2 n2
= qn
H qK N qQ  q n
q 2 n2

E(X2) = n(n – 1) p2 + np = (n2 – n) P2 + np

= n2 p2 – np2 + np = [np – p + 1]
= np[np + (1 – p)] = np(np + q)
= n2 p2 + npq
 Variance = (n2p2 + npq) – (np)2 = npq

Poisson Distribution

The Poisson distribution is used to describe discrete quantitative data such as counts
in which the populations size n is large, the probability of an individual event  is small,
but the expected number of events, n  is moderate say five or more.
Typical examples are the number of deaths in a town from a particular disease per
day or the number of admissions to a particular hospital.
Poisson distribution is a discrete probability distribution developed by French
Mathematician, S. Poisson. Poisson distribution is used when the chance of any individual
event being a success is small. The distribution is given by the following formula.

e m mr
P(r) n =
r!
Where r = 0, 1, 2, 3, 4, ………
e = 2.7183 (the base of natural logarithms)
m = mean of poisson distribution = np
(average no. of occurrences of an event)
The poisson distribution is a discrete distribution with a single parameter ‘m’. As
‘m’ increases, the distribution shifts to the right. All Poisson probability distribution are
skewed to right. Poisson distribution is a distribution of rate events (the probabilities tend
to be high for small numbers of occurrences). Mean = variance = m (for a Poisson
distribution).

Amity Directorate of Distance and Online Education

176 Quantitative Techniques in Management

Notes Illustration - 5

Suppose that on an average 1 house in 1000 in a certain district has a fire during
a year. If there are 3000 houses in that districts, What is the probability that exactly 4
houses will have a fire during the year?

Solution:

n = No. of houses = 3,000

p = Probability of house catching fire

1
= = 0.001
1000
Since n is large and p is small, Poisson distribution is applied.
Mean or expected number of houses to each fire is
m = np = 3,000 × 0.001 = 1
Probability that ‘r’ occurrences will take place in ‘n’ is

e m mr
p(r) =
r!
Probability that exactly 4 houses will have a fire during the year is

e 3 m4 0.050  81
p(4) = = = 0.169
4! 24

Illustration - 6

In a Poisson distribution, the probability P(x) for X = 0 is 10 percent. Find the mean
of the distribution.
Solution:
By Poisson distribution,

e m mr
P(r) =
r!
where, m mean of the distribution
Given P(0) = 0.10 (i.e. 10%)

e m  1 e m  1
 P(0 = = 0.1  = 0.1  e- m = 0.1
1 1
 m = 2.3026
(To obtain ‘m’, take natural log of 0.1)
Mean of the distribution = 2.3026.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 177

Normal Distribution Notes

The probability distribution used mostly for dealing with quantities whose magnitude
is continuously variable is normal distribution. It is also called normal probability
distribution. The distribution plays a very important and pivotal role in statistical theory
and practices, particularly in the area of statistical inference. Its important is also due
to the fact that in practice, the experiential results very often, seem to follow the normal
distribution or the ‘bell-shaped curve’. The normal curve is symmetrical and is defined by
its mean  and standard deviation  .
The normal distribution is approximation to binomial distribution. Whether or not p
= q, the binomial distribution tends to the form of a continuous curve and when ‘n’ becomes
large at least for the material part of the range. The normal distribution is defined as:

1
P(x) =
 2
Where, x = Values of continuous random variable
 = Mean of the normal random variable
E = Mathematical constant = 2.7183
 = Mathematical constant = 3.1416
 = Standard deviation of normal random.

Properties of Normal Distribution

i) The normal curve is ‘bell-shaped’ and symmetrical in its appearance. If the
curves were folded along its vertical axis, the two halves would coincide. The
number of cases below the mean is equal to that above the mean.
ii) The height of the normal curve is at maximum at the mean. Hence, the mean
and mode of the normal distribution coincide. Thus, for a normal distribution,
mean = mode = median.
iii) There is one maximum point curve which occurs at the mean. The height of
the curve declines as it go in either direction from the mean. The curve
approaches nearer and nearer to the base but it never touches it i.e., the curve
is asymptotic.
iv) Since there is only one maximum point, the normal curve is unimodal, i.e., it
has only one mode.
v) The points of inflexion i.e., where the change in curveature occurs are X 
.
vi) The first and third quartiles are equidistant from the median.
vii) The variable distributed according to the normal curve is a continuous one.
viii) The area under the normal curve is distributed as follows.
a) Mean  1  covers 68.27% area.
b) Mean  2  covers 95.45% area.
c) Mean  3  covers 99.73% area.
It is often the case with medical data that the histogram of a continuous variable
obtained from a single measurement on different subjects will have a characteristic ‘bell-

Amity Directorate of Distance and Online Education

178 Quantitative Techniques in Management

Notes shaped’ distribution known as a Normal distribution. One such example is the histogram
of the birth weight in kilograms of the 3,226 new born babies.

Fig: Distribution of birth weight in 3,226 newborn babies

(data from O' Cathain et al 2002)
To distinguish the use of the same word in normal range and Normal distribution.
The histogram of the sample data is an estimate of the population distribution of
birth weights in new born babies. This population distribution can be estimated by the
superimposed smooth `bell-shaped' curve or ‘Normal’ distribution shown. The entire
population of new born babies then the distribution of birth weight would have exactly the
Normal shape. The inference, from a sample whose histogram has the approximate Normal
shape, that the population will have exactly or as near as makes no practical difference,
that Normal shape.
The Normal distribution is completely described by two parameters µ and s, where
µ represents the population mean or centre of the distribution and s the population standard
deviation. Populations with small values of the standard deviation s have a distribution
concentrated close to the centre µ; those with large standard deviation have a distribution
widely spread along the measurement axis. One mathematical property of the Normal
distribution is that exactly 95% of the distribution lies between
µ - (1.96   ) and µ + (1.96   )
Changing the multiplier 1.96 to 2.58, exactly 99% of the Normal distribution lies in
the corresponding interval.
In practice the two parameters of the Normal distribution, µ and s, must be estimated
from the sample data.
For this purpose a random sample from the population is first taken. The sample
mean x and the sample standard deviation, SD( x ) = s, are then calculated. If a sample
is taken from such a Normal distribution and provided the sample is not too small, then
approximately 95% of the sample will be covered by

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 179

x  [1.96  SD(x)] to x + [1.96  SD(x)] Notes

This is calculated by merely replacing the population parameters µ and s by the
sample estimates x and s in the previous expression.
In appropriate circumstances this interval may estimate the reference interval for a
particular laboratory test which is then used for diagnostic purposes.
Normally distributed to calculate a reference range. We have already mentioned that
about 95% of the observations lie within +/-1.96SDs of the mean. So a reference range
for our sample of babies is:
3.39  [1.96  0.55] to 3.39 + [1.96  0.55]
2.31kg to 4.47kg
A baby's weight at birth is strongly associated with mortality risk during the first year
and, to a lesser degree, with developmental problems in childhood and the risk of various
diseases in adulthood. If the data are not normally distributed then we can base the normal
reference range on the observed percentiles of the sample. i.e., 95% of the observed data
lie between the 2.5 and 97.5 percentiles.
So a percentile-based reference range is: 2.19 kg to 4.43 kg.
Most reference ranges are based on samples larger than 3500 people. Over many
years and millions of births, the WHO has come up with a with a normal birth weight
range for new born babies. These ranges represent results than are acceptable in new
born babies and actually cover the middle 80% of the population distribution i.e. the 10th
and 90th centiles. Low birth weight babies are usually defined by the WHO is as weighing
less than 2500g regardless of gestational age, since and large birth weight babies are
defined as weighing above 4000 kg.
Hence the normal birth weight range is around 2.5 kg to 4 kg. For sample data,
the 10 to 90th centile range was similar, 2.75 to 4.03 kg.

Illustration - 1

1000 light bulbs with a mean life of 120 days are installed in a firm. Their length
of life is normally distributed with a standard deviation of 20 days. How many bulbs will
expire in less than 90 days?

Solution:

No. of bulbs expiring in less than 90 days.

xx 90  120 30

When x = 90, z  z    1.5
 20 20

Amity Directorate of Distance and Online Education

180 Quantitative Techniques in Management

Notes x = 90 x = 120
z = 1.5 z = 0
Area between Z = 0 and Z = -1.5 is 0.4332
 Area less than 90 days = 0.5 - 0.4332 = 0.668
Therefore the number of bulbs that will expire in loess than 90 days is 0.0668 x 1000
= 66.8 bulbs.

Illustration - 2

The mean circumference of 1500 shafts manufactured in a company is 15 cm and

the deviation from the mean is 3 cm. Assuming normal distribution find out how many
shafts have a circumference.
a) Greater than 13 cm
b) Lesser then 19 cm
Solution:
Given, SD (  ) = 3 and Mean (  ) = 15 and n = 1,500

X  15
Then, Z  is N (0,1)
3

 x  15 13  15 
a) P(x > 13) = P   = P(Z> – 0.666)
 3 3 

 -0.666 
= area from –0.666 to  = 0.7454

 x  15 19  15 
b) P(x < 19) = P   = P(Z > 1.33)
 3 3 
= (area from –  (0  ) – (area from 1.33 to  )
= 1 – 0.1515 = 0.8485

Hypergeometric Distribution

The hypergeometric distribution is a probability distribution that’s very similar to the

binomial distribution. In fact, the binomial distribution is a very good approximation of the
hypergeometric distribution as long as you are sampling 5% or less of the population.
A hypergeometric random variable is  the  number  of  successes that  result from a
hypergeometric experiment. The probability distribution of a hypergeometric random
variable is called a hypergeometric distribution.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 181

The hyper geometric distribution has the following characteristics: Notes

(i) There are only 2 possible outcomes.
(ii) The probability of a success is not the same on each trial.
(iii) It results from a count of the number of successes in a fixed number of trials.
The definition for the hypergeometric distribution, where X is a random variable, is:

 K  N  K 
  
 k  n  k 
P ( X  k)
N 
 
n 

Where, K is the number of successes in the population. k is the number of observed

successes. N is the population size. n is the number of draws.
Example:
A deck of cards contains 20 cards: 6 red cards and 14 black cards. 5 cards are
drawn randomly without replacement. What is the probability that exactly 4 red cards are
drawn?
The probability of choosing exactly 4 red cards is: P(4 red cards) = # samples with
4 red cards and 1 black card / # of possible 4 card samples.
Using the combinations formula, the problem becomes:

 6 14 
  
4 1
P ( X  k )   
 20 
 
5 

The above formula can be written as: C

6
4 
 14C1  20
C5 where 6C4 means that
14
out of 6 possible red cards, we are choosing 4. C1 means that out of a possible 14
black cards, we’re choosing 1.

Solution:

14
C6
4 
 14C1  20
C5  15 
15504
= 0.0135

The binomial distribution doesn’t apply here, because the cards are not replaced once
they are drawn. In other words, the trials are not independent events. For example, for
1 red card, the probability is 6/20 on the first draw. If that card is red, the probability of
choosing another red card falls to 5/19.
When given x, N, n and k, we can compute the hypergeometric probability based on
the following formula:
The population consists of N items, k of which are successes. And a random sample
drawn from that population consists of nitems, x of  which  are  successes.  Then  the
hypergeometric probability is:
h(x; N, n, k) = [ kCx ]  [ N-kCn-x ]  /  [ NCn ]

Amity Directorate of Distance and Online Education

182 Quantitative Techniques in Management

Notes The hypergeometric distribution has the following properties:

(i) The mean of the distribution is equal to n * k / N .
(ii) The variance is n * k *  ( N - k )  *  ( N - n )  /  [ N2 *  ( N -  1  )  ]  .

Exponential Distribution

Another useful continuous distribution is the exponential distribution, which has the
following probability density function:

f(x) = λe  λx For, x  0.

This family of distributions is characterized by a single parameter λ, which is called

the rate. Intuitively, λ can be thought of as the instantaneous “failure rate” of a “device”
at any time t, given that the device has survived up to t.
The exponential distribution is typically used to model time intervals between “random
events”.
Example:
1. The length of time between telephone calls.
2. The length of time between arrivals at a service station.
3. The life time of electronic components, i.e., an inter-failure time.
An important fact is that when times between random “events” follow the exponential
distribution with rate λ, then the total number of events in a time period of length t follows
the Poisson distribution with parameter?
If a random variable X is exponentially distributed with rate ë, then it can be shown
that: E(X) = 1/λ and V (X) = (1/λ)2 .

For, λ = 0.5, 1 and 2, the shapes of the exponential density curve are:
Observe that the greater the rate, the faster the curve drops. Or, the lower the rate,
the flatter the curve.
Several useful formulas are:
-λx
P{X  x} = 1 - e

P{X > x} = e -λx

-λx -λx
P{x1 < X  x2} = e 1-e
These correspond to the areas under the density curve to the left of x, to the right
of x and between x1 and x2, respectively.

4.5 Sampling
Sampling is an important concept which is practiced in every activity. Sampling
involves selecting a relatively small number of elements from a large defined group of
elements and expecting that the information gathered from the small group will allow
judgments to be made about the large group. The basic idea of sampling is that by
selecting some of the elements in a population, the conclusion about the entire population
is drawn. Sampling is used when conducting census is impossible or unreasonable. In

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 183

a census method a researcher collects primary data from every member of a defined target Notes
population. It is not always possible or necessary to collect data from every unit of the
population. The researcher can resort to sample survey to find answers to the research
questions. However, they can do more harm than good if the data is not collected from
the people, events or objects that can provide correct answers to the problem. The process
of selecting the right individuals, objects or events for the purpose of the study is known
as sampling

Meaning of Sampling

Sampling is defined as the selection of some part of an aggregate or totality on the

basis of which a judgment or inference about the aggregate or totality is made. Sampling
is the process of learning about the population on the basis of a sample drawn from it.

Purpose of Sampling

There are several reasons for sampling. They are explained below:
(i) Lower cost: The cost of conducting a study based on a sample is much lesser
than the cost of conducting the census study.
(ii) Greater accuracy of results: It is generally argued that the quality of a study
is often better with sampling data than with a census. Research findings also
substantiate this opinion.
(iii) Greater speed of data collection: Speed of execution of data collection is
higher with the sample. It also reduces the time between the recognition of a
need for information and the availability of that information.
(iv) Availability of population element: Some situations require sampling. When
the breaking strength of materials is to be tested, it has to be destroyed. A
census method cannot be resorted as it would mean complete destruction of
all materials. Sampling is the only process possible if the population is infinite.

Features of Sampling Method

The sampling technique has the following good features of value and significance:
1. Economy: Sampling technique brings about cost control of a research project
as it requires much less physical resources as well as time than the census
technique.
2. Reliability: In sampling technique, if due diligence is exercised in the choice
of sample unit and if the research topic is homogenous then the sample survey
can have almost the same reliability as that of census survey.
3. Detailed Study: An intensive and detailed study of sample units can be done
since their number is fairly small. Also multiple approaches can be applied to
a sample for an intensive analysis.
4. Scientific Base: As mentioned earlier this technique is of scientific nature as
the underlined theory is based on principle of statistics.
5. Greater Suitability in most Situations: It has a wide applicability in most
situations as the examination of few sample units normally suffices.
6. Accuracy: The accuracy is determined by the extent to which bias is eliminated
from the sampling. When the sample elements are drawn properly some sample
elements underestimates the population values being studied and others
overestimate them.

Amity Directorate of Distance and Online Education

184 Quantitative Techniques in Management

Notes Essentials of Sampling

In order to reach a clear conclusion, the sampling should possess the following
essentials:
1. It must be representative: The sample selected should possess the similar
characteristics of the original universe from which it has been drawn.
2. Homogeneity: Selected samples from the universe should have similar nature
and should not have any difference when compared with the universe.
3. Adequate Samples: In order to have a more reliable and representative result,
a good number of items are to be included in the sample.
4. Optimization: All efforts should be made to get maximum results both in terms
of cost as well as efficiency. If the size of the sample is larger, there is better
efficiency and at the same time the cost is more. A proper size of sample is
maintained in order to have optimized results in terms of cost and efficiency.

4.6 Steps in Developing a Sampling Plan

A number of concepts, procedures and decisions must be considered by a researcher
in order to successfully gather raw data from a relatively small group of people which in
turn can be used to generalize or make predications about all the elements in a larger
target population.
The following are the logical steps involved in the sample execution:
1) Define the target population
The first task of a researcher is to determine and identify the complete group of people
or objects that should be included in the study. With the statement of the problem and
the objectives of the study acting as guideline the target population should be identified
on the basis of descriptors that represent the characteristics features of element that make
the target population’s frame. These elements become the prospective sampling unit from
which a sample will be drawn. A clear understanding of the target population will enable
the researcher to successfully draw a representative sample.
2) Select the data collection method
Based on the problem definition, the data requirements and the research objectives,
the researcher should select a data collection method for collecting the required data from
the target population elements. The method of data collection guides the researcher in
identifying and securing the necessary sampling frame for conducting the research.
3) Identify the sampling frames needed
The researcher should identify and assemble a list of eligible sampling units. The
list should contain enough information about each prospective sampling unit so as to
enable the researcher to contact them. Drawing an incomplete frame decreases the
likelihood of drawing a representative sample.
4) Select the appropriate sampling method
The researcher can choose between probability and non-probability sampling
methods. Using a probability sampling method will always yield better and more accurate
information about the target population’s parameters than the non-probability sampling
methods. Seven factors should be considered in deciding the appropriateness of the
sampling method viz., research objectives, degree of desired accuracy, availability of

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 185

resources, time frame, advanced knowledge of the target population, scope of the research Notes
and perceived statistical analysis needs.
5) Determine necessary sample sizes and overall contact rates
The sample size is decided based on the precision required from the sample
estimates, time and money available to collect the required data. While determining the
sample size due consideration should be given to the variability of the population
characteristic under investigation, the level of confidence desired in the estimates and the
degree of the precision desired in estimating the population characteristic. The number
of prospective units to be contacted to ensure that the estimated sample size is obtained
and the additional cost involved should be considered. The researcher should calculate
the reachable rates, overall incidence rate and expected completion rates associated with
the sampling situation.
6) Creating an operating plan for selecting sampling units
The actual procedure to be used in contacting each of the prospective respondents
selected to form the sample should be clearly laid out. The instruction should be clearly
written so that interviewers know what exactly should be done and the procedure to be
followed in case of problems encountered, in contacting the prospective respondents.
7) Executing the operational plan
The sample respondents are met and actual data collection activities are executed
in this stage. Consistency and control should be maintained at this stage.

4.7 Sampling Techniques

The sampling design can be broadly grouped on two basis viz., representation and
element selection. Representation refers to the selection of members on a probability or
by other means. Element selection refers to the manner in which the elements are selected
individually and directly from the population. If each element is drawn individually from the
population at large, it is an unrestricted sample. Restricted sampling is where additional
controls are imposed, in other words it covers all other forms of sampling.
The classification of sampling design on the basis of representation and element
selection is shown below:

I. Probability Sampling

Probability sampling is where each sampling unit in the defined target population
has a known non-zero probability of being selected in the sample. The actual probability
of selection for each sampling unit may or may not be equal depending on the type of
probability sampling design used. Specific rules for selecting members from the operational
population are made to ensure unbiased selection of the sampling units and proper sample
representation of the defined target population. The results obtained by using probability
sampling designs can be generalized to the target population within a specified margin
of error.
Probability samples are characterised by the fact that, the sampling units are
selected by chance. In such a case, each member of the population has a known, non-
zero probability of being selected. However, it may not be true that all samples would
have the same probability of selection, but it is possible to say the probability of selecting
any particular sample of a given size. It is possible that one can calculate the probability
that any given population element would be included in the sample. This requires a precise
definition of the target population as well as the sampling frame. Probability sampling
Amity Directorate of Distance and Online Education
186 Quantitative Techniques in Management

Notes techniques differ in terms of sampling efficiency which is a concept that refers to trade
off between sampling cost and precision. Precision refers to the level of uncertainty about
the characteristics being measured. Precision is inversely related to sampling errors but
directly related to cost. The greater the precision, the greater the cost and there should
be a trade-off between sampling cost and precision. The researcher is required to design
the most efficient sampling design in order to increase the efficiency of the sampling.
The different types of probability sampling designs are discussed below:
1) Simple Random Sampling

The following are the implications of random sampling:

(i) It provides each element in the population an equal probability chance of being
chosen in the sample, with all choices being independent of one another and
(ii) It offers each possible sample combination an equal probability opportunity of
being selected.
In the unrestricted probability sampling design every element in the population has
a known, equal non-zero chance of being selected as a subject. For example, if 10
employees (n = 10) are to be selected from 30 employees (N = 30), the researcher can
write the name of each employee in a piece of paper and select them on a random basis.
Each employee will have an equal known probability of selection for a sample. The same
is expressed in terms of the following formula:
Probability of selection = Size of sample / Size of population
Each employee would have a 10/30 or .333 chance of being randomly selected in
a drawn sample. When the defined target population consists of a larger number of
sampling units, a more sophisticated method can be used to randomly draw the necessary
sample. A table of random numbers can be used for this purpose. The table of random
numbers contains a list of randomly generated numbers. The numbers can be randomly
generated through the computer programs also. Using the random numbers the sample
can be selected.
Advantages and Disadvantages
The simple random sampling technique can be easily understood and the survey
result can be generalized to the defined target population with a pre specified margin of
error. It also enables the researcher to gain unbiased estimates of the population’s
characteristics. The method guarantees that every sampling unit of the population has
a known and equal chance of being selected, irrespective of the actual size of the sample
resulting in a valid representation of the defined target population.
The major drawback of the simple random sampling is the difficulty of obtaining
complete, current and accurate listing of the target population elements. Simple random
sampling process requires all sampling units to be identified which would be cumbersome
and expensive in case of a large population. Hence, this method is most suitable for a
small population.

2) Systematic Random Sampling

The systematic random sampling design is similar to simple random sampling but
requires that the defined target population should be selected in some way. It involves
drawing every nth element in the population starting with a randomly chosen element
between 1 and n. In other words individual sampling units are selected according their
position using a skip interval. The skip interval is determined by dividing the sample size

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 187

into population size. For example, if the researcher wants a sample of 100 to be drawn Notes
from a defined target population of 1000, the skip interval would be 10(1000/100). Once
the skip interval is calculated, the researcher would randomly select a starting point and
take every 10th until the entire target population is proceeded through. The steps to be
followed in a systematic sampling method are enumerated below:
(i) Total number of elements in the population should be identified
(ii) The sampling ratio is to be calculated ( n = total population size divided by
size of the desired sample)
(iii) A sample can be drawn by choosing every nth entry
Two important considerations in using the systematic random sampling are:
(i) It is important that the natural order of the defined target population list be
unrelated to the characteristic being studied.
(ii) Skip interval should not correspond to the systematic change in the target
population.
Advantages and Disadvantages
The major advantage is its simplicity and flexibility. In case of systematic sampling
there is no need to number the entries in a large personnel file before drawing a sample.
The availability of lists and shorter time required to draw a sample compared to random
sampling makes systematic sampling an attractive, economical method for researchers.
The greatest weakness of systematic random sampling is the potential for the hidden
patterns in the data that are not found by the researcher. This could result in a sample
not truly representative of the target population. Another difficulty is that the researcher
must know exactly how many sampling units make up the defined target population. In
situations where the target population is extremely large or unknown, identifying the true
number of units is difficult and the estimates may not be accurate.

3) Stratified Random Sampling

Stratified random sampling requires the separation of defined target population into
different groups called strata and the selection of sample from each stratum. Stratified
random sampling is very useful when the divisions of target population are skewed or when
extremes are present in the probability distribution of the target population elements of
interest. The goal in stratification is to minimize the variability within each stratum and
maximize the difference between strata. The ideal stratification would be based on the
primary variable under study. Researchers often have several important variables about
which they want to draw conclusions. A reasonable approach is to identify some basis
for stratification that correlates well with other major variables. It might be a single variable
like age, income etc. or a compound variable like on the basis of income and gender.
Stratification leads to segmenting the population into smaller, more homogeneous sets
of elements. In order to ensure that the sample maintains the required precision in terms
of representing the total population, representative samples must be drawn from each of
the smaller population groups.
There are three reasons as to why a researcher chooses a stratified random sample:
(i) To increase the sample’s statistical efficiency
(ii) To provide adequate data for analyzing various sub populations
(iii) To enable different research methods and procedures to be used in different
strata.

Amity Directorate of Distance and Online Education

188 Quantitative Techniques in Management

Notes Drawing a stratified random sampling involves the following steps:

1. Determine the variables to use for stratification
2. Select proportionate or disproportionate stratification
3. Divide the target population into homogeneous subgroups or strata
4. Select random samples from each stratum
5. Combine the samples from each stratum into a single sample of the target
population.
There are two common methods for deriving samples from the strata viz.,
proportionate and disproportionate. In proportionate stratified sampling, each stratum is
properly represented so the sample drawn from it is proportionate to the stratum’s share
of the total population. The larger strata are sampled more because they make up a larger
percentage of the target population. This approach is more popular than any other stratified
sampling procedures due to the following reasons:
(i) It has higher statistical efficiency than the simple random sample
(ii) It is much easier to carry out than other stratifying methods
(iii) It provides a self-weighing sample i.e., the population mean or proportion can
be estimated simply by calculating the mean or proportion of all sample cases.
In disproportionate stratified sampling, the sample size selected from each stratum
is independent of that stratum’s proportion of the total defined target population. This
approach is used when stratification of the target population produces sample sizes that
contradict their relative importance to the study. An alternative of disproportionate stratified
method is optimal allocation. In this method, consideration is given to the relative size
of the stratum as well as the variability within the stratum to determine the necessary
sample size of each stratum. The logic underlying the optimal allocation is that the greater
the homogeneity of the prospective sampling units within a particular stratum, the fewer
the units that would have to be selected to estimate the true population parameter
accurately for that subgroup. This method is also opted for in situation where it is easier,
simpler and less expensive to collect data from one or more strata than from others.
Stratified random sampling provides several advantages viz., the assurance of
representativeness in the sample, the opportunity to study each stratum and make relative
comparisons between strata and the ability to make estimates for the target population
with the expectation of greater precision or less error.

4) Cluster Sampling

Cluster sampling is a probability sampling method in which the sampling units are
divided into mutually exclusive and collectively exhaustive subpopulation called clusters.
Each cluster is assumed to be the representative of the heterogeneity of the target
population. Groups of elements that would have heterogeneity among the members within
each group are chosen for study in cluster sampling. Several groups with intragroup
heterogeneity and intergroup homogeneity are found. A random sampling of the clusters
or groups is done and information is gathered from each of the members in the randomly
chosen clusters. Cluster sampling offers more of heterogeneity within groups and more
homogeneity among the groups.
Single Stage and Multistage Cluster Sampling
In single stage cluster sampling, the population is divided into convenient clusters
and required number of clusters are randomly chosen as sample subjects. Each element

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 189

in each of the randomly chosen cluster is investigated in the study. Cluster sampling can Notes
also be done in several stages which is known as multistage cluster sampling. For
example: To study the banking behaviour of customers in a national survey, cluster
sampling can be used to select the urban, semi-urban and rural geographical locations
of the study. At the next stage, particular areas in each of the location would be chosen.
At the third stage, the banks within each area would be chosen. Thus multi-stage sampling
involves a probability sampling of the primary sampling units; from each of the primary
units, a probability sampling of the secondary sampling units is drawn; a third level of
probability sampling is done from each of these secondary units, and so on until the final
stage of breakdown for the sample units are arrived at, where every member of the unit
will be a sample.
Area Sampling
Area sampling is a form of cluster sampling in which the clusters are formed by
geographic designations. For example, state, district, city, town etc., Area sampling is
a form of cluster sampling in which any geographic unit with identifiable boundaries can
be used. Area sampling is less expensive than most other probability designs and is not
dependent on population frame. A city map showing blocks of the city would be adequate
information to allow a researcher to take a sample of the blocks and obtain data from
the residents therein.
Advantages and Disadvantages of Cluster Sampling
The cluster sampling method is widely used due to its overall cost-effectiveness and
feasibility of implementation. In many situations the only reliable sampling unit frame
available to researchers and representative of the defined target population, is one that
describes and lists clusters. The list of geographical regions, telephone exchanges, or
blocks of residential dwelling can normally be easily compiled than the list of all the
individual sampling units making up the target population. Clustering method is a cost
efficient way of sampling and collecting raw data from a defined target population.
One major drawback of clustering method is the tendency of the cluster to be
homogeneous. The greater the homogeneity of the cluster, the less precise will be the
sample estimate in representing the target population parameters. The conditions of intra-
cluster heterogeneity and inter-cluster homogeneity are often not met. For these reasons
this method is not practiced often.
5) Sequential/Multiphase Sampling
This is also called Double Sampling. Double sampling is opted when further
information is needed from a subset of groups from which some information has already
been collected for the same study. It is called as double sampling because initially a
sample is used in the study to collect some preliminary information of interest and later
a sub-sample of this primary sample is used to examine the matter in more detail The
process includes collecting data from a sample using a previously defined technique.
Based on this information, a sub sample is selected for further study. It is more convenient
and economical to collect some information by sampling and then use this information
as the basis for selecting a sub sample for further study.
6) Sampling with Probability Proportional to Size
When the case of cluster sampling units does not have exactly or approximately
the same number of elements, it is better for the researcher to adopt a random selection
process, where the probability of inclusion of each cluster in the sample tends to be
proportional to the size of the cluster. For this, the number of elements in each cluster
has to be listed, irrespective of the method used for ordering it. Then the researcher should

Amity Directorate of Distance and Online Education

190 Quantitative Techniques in Management

Notes systematically pick the required number of elements from the cumulative totals. The actual
numbers thus chosen would not however reflect the individual elements, but would indicate
as to which cluster and how many from them are to be chosen by using simple random
sampling or systematic sampling. The outcome of such sampling is equivalent to that
of simple random sample. This method is also less cumbersome and is also relatively
less expensive.

II. Non-probability Sampling

In non probability sampling method, the elements in the population do not have any
probabilities attached to being chosen as sample subjects. This means that the findings
of the study cannot be generalized to the population. However, at times the researcher
may be less concerned about generalizability and the purpose may be just to obtain some
preliminary information in a quick and inexpensive way. Sometimes when the population
size is unknown, then non probability sampling would be the only way to obtain data.
Some non-probability sampling techniques may be more dependable than others and could
often lead to important information with regard to the population.
Non-probability sampling does not involve random selection. It involves personal
judgement of the researcher rather than chance to select sample elements. Sometimes
this judgement is imposed by the researcher, while in other cases the selection of
population elements to be included is left to the individual field workers. The decision maker
may also contribute to including a particular individual in the sampling frame. Evidently,
non probability sampling does not include elements selected probabilistically and hence,
leaves a degree of sampling error associated with the sample. Sampling error is the degree
to which a sample might differ from the population. Therefore, while inferring to the
population, results could not be reported plus or minus the sampling error. In non-
probability sampling, the degree to which the sample differs from the population remains
Unknown However, we cannot come to a conclusion that sampling error is an inherent
of non probability sample. on-probability samples also yield good estimates of the
population characteristics. Since, inclusion of the elements in the sample are not
determined in a probabilistic way, the estimates obtained are not statistically projectable
to the population.

1) Convenience Sampling

Non-probability samples that are unrestricted are called convenient sampling.

Convenience sampling refers to the collection of information from members of population
who are conveniently available to provide it. Researchers or field workers have the freedom
to choose as samples whomever they find, thus it is named as convenience. It is mostly
used during the exploratory phase of a research project and it is the best way of getting
some basic information quickly and efficiently. The assumption is that the target population
is homogeneous and the individuals selected as samples are similar to the overall defined
target population with regard to the characteristics being studied. However, in reality there
is no way to accurately assess the representativeness of the sample. Due to the self
selection and voluntary nature of participation in data collection process the researcher
should give due consideration to the non-response error.
Advantages and Disadvantages
Convenient sampling allows a large number of respondents to be interviewed in a
relatively short time. This is one of the main reasons for using convenient sampling in
the early stages of research. However the major drawback is that the use of convenience

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 191

samples in the development phases of constructs and scale measurements can have a Notes
serious negative impact on the overall reliability and validity of those measures and
instruments used to collect raw data. Another major drawback is that the raw data and
results are not generalizable to the defined target population with any measure of precision.
It is not possible to measure the representativeness of the sample, because sampling
error estimates cannot be accurately determined.

2) Judgment Sampling

Judgment sampling is a non-probability sampling method in which participants are

selected according to an experienced individual’s belief that they will meet the
requirements of the study. The researcher selects sample members who conform to some
criterion. It is appropriate in the early stages of an exploratory study and involves the choice
of subjects who are most advantageously placed or in the best position to provide the
information required. This is used when a limited number or category of people have the
information that are being sought. The underlying assumption is that the researcher’s belief
that the opinions of a group of perceived experts on the topic of interest are representative
of the entire target population.
Advantages and Disadvantages
If the judgment of the researcher or expert is correct then the sample generated from
the judgment sampling will be much better than one generated by convenience sampling.
However, as in the case of all non-probability sampling methods, the representativeness
of the sample cannot be measured. The raw data and information collected through
judgment sampling provides only a preliminary insight.

3) Quota Sampling

The quota sampling method involves the selection of prospective participants

according to pre specified quotas regarding either the demographic characteristics (gender,
age, education, income, occupation etc.,) specific attitudes (satisfied, neutral, dissatisfied)
or specific behaviours (regular, occasional, rare user of product). The purpose of quota
sampling is to provide an assurance that pre specified subgroups of the defined target
population are represented on pertinent sampling factors that are determined by the
researcher. It ensures that certain groups are adequately represented in the study through
the assignment of the quota.
Advantages and Disadvantages
The greatest advantage of quota sampling is that the sample generated contains
specific subgroups in the proportion desired by researchers. In those research projects
that require interviews the use of quotas ensures that the appropriate subgroups are
identified and included in the survey. The quota sampling method may eliminate or reduce
selection bias.
An inherent limitation of quota sampling is that the success of the study will be
dependent on subjective decisions made by the researchers. As a non-probability method,
it is incapable of measuring true representativeness of the sample or accuracy of the
estimate obtained. Therefore, attempts to generalize the data results beyond those
respondents who were sampled and interviewed become very questionable and may
misrepresent the given target population.

Amity Directorate of Distance and Online Education

192 Quantitative Techniques in Management

Notes 4) Snowball Sampling

Snowball sampling is a non-probability sampling method in which a set of

respondents are chosen who help the researcher to identify additional respondents to be
included in the study. This method of sampling is also called as referral sampling because
one respondent refers other potential respondents. This method involves probability and
non-probability methods. The initial respondents are chosen by a random method and the
subsequent respondents are chosen by non-probability methods. Snowball sampling is
typically used in research situations where the defined target population is very small and
unique and compiling a complete list of sampling units is a nearly impossible task. This
technique is widely used in academic research. While the traditional probability and other
non-probability sampling methods would normally require an extreme search effort to
qualify a sufficient number of prospective respondents, the snowball method would yield
better result at a much lower cost. The researcher has to identify and interview one qualified
respondent and then solicit his help to identify other respondents with similar
characteristics.
Advantages and Disadvantages
Snowball sampling enables to identify and select prospective respondents who are
small in number, hard to reach and uniquely defined target population. It is most useful
in qualitative research practices. Reduced sample size and costs are the primary
advantage of this sampling method. The major drawback is that the chance of bias is
higher. If there is a significant difference between people who are identified through snowball
sampling and others who are not then, it may give rise to problems. The results cannot
be generalized to members of larger defined target population.

4.8 Merits and Limitations of Sampling

Merits of Sampling

1. It saves time: Sampling method of data collection saves time because fewer
items are collected and processed. When the results are urgently required, this
method is very helpful.
2. It reduces cost: Since only a few and selected items are studied in sampling,
there is reduction in cost of money and reduction in terms of man hours.
3. More reliable results can be obtained: Through sampling, more reliable results
can be obtained because (a) there are fewer chances of sampling statistical
errors. If there is sampling error, it is possible to estimate and control the results.
(b) Highly experienced and trained persons can be employed for scientific
processing and analyzing of relatively limited data and they can use their high
technical knowledge and get more accurate and reliable results.
4. It provides more detailed information: As it saves time, money and labor,
more detailed information can be collected in a sample survey.
5. Dependency on Sampling method: Some times it so happens that one has
to depend upon sampling method alone because if the population under study
is finite, sampling method is the only method to be used. For example, if
someone’s blood has to be examined, it will become fatal to take all the blood
out from the body and study depending upon the total enumeration method.
6. Administrative convenience: The organization and administration of sample
survey are easy and convenient for the researcher.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 193

7. More scientific: Since the methods used to collect data are based on scientific Notes
theory and results obtained can be tested, sampling is a more scientific method
of collecting data.

Limitations of Sampling
1. Less Accuracy: In comparison to census technique the conclusions derived
from samples are more liable to error. Therefore, sampling technique is less
accurate than the census technique.
2. Changeability of Units: If the units in the field of survey are liable to change
or if these are not homogenous the sampling technique will be very hazardous.
It is not scientific to extend the conclusions derived from one set of sample
to other sets which are dissimilar or are changeable.
3. Misleading Conclusions: If due care is not taken in the selection of samples
or if they are arbitrarily selected, the conclusions derived from them will become
misleading if extended to all units. For example, in assessing the monthly
expenditure of university students we select a sample study of only rich
students, our results will be highly erroneous if extended to all students.
4. Need for Specialized Knowledge: The sample technique can be successful
only if a competent and able researcher makes the selection. If this is done
by an average researcher, the selection is liable to be wrong.
5. Situational Impact: Under certain circumstances it is very difficult to use the
sampling technique. If the time is very short and it is not possible to make
selection of the sample, the technique cannot be used. Besides, if we need
100% accuracy the sampling technique cannot be used. It can also not be used
if the material is of heterogeneous nature.
6. Sample not representative: To make the sample representative is a difficult
task. If a representative sample is taken from the universe, the result is
applicable to the whole population. If the sample is not representative of the
universe the result may be false and misleading.
7. Complicated Process: Sometimes the sampling plan may be complicated and
requires more money, labor and time than a census method.
8. Personal bias: There may be personal biases and prejudices with regard to
the choice of technique and drawing of sampling units on the part of the
researcher.

4.9 Sampling Distribution

A sample is that part of the universe which the select for the purpose of investigation.
A sample exhibits the characteristics of the universe. The word sample literally means
small universe. For example, suppose the microchips produced in a factory are to be
tested. The aggregate of all such items is universe, but it is not possible to test every
item. So in such a case, a part of the universe is taken and then tested. Now this quantity
extracted for testing is known as sample.
If we take certain number of samples and for each sample compute various statistical
measures such as mean, standard deviation etc. then we can find out that each sample
may give its own value for statistics under consideration. All such values of a particular
statics, say, mean together with their relative frequencies will constitute the sampling
distribution of mean standard deviation.

Amity Directorate of Distance and Online Education

194 Quantitative Techniques in Management

Notes 4.10 Hypothesis

Hypothesis test is a method of making decisions using data from a scientific study.
In statistics, a result is called statistically significant if it has been predicted as unlikely
to have occurred by chance alone, according to a pre-determined threshold probability,
the significance level. The phrase “test of significance” was coined by statistician Ronald
Fisher. These tests are used in determining what outcomes of a study would lead to a
rejection of the null hypothesis for a pre-specified level of significance; this can help to
decide whether results contain enough information to cast doubt on conventional wisdom,
given that conventional wisdom has been used to establish the null hypothesis. The critical
region of a hypothesis test is the set of all outcomes which cause the null hypothesis
to be rejected in favor of the alternative hypothesis. Statistical hypothesis testing is
sometimes called confirmatory data analysis, in contrast to exploratory data analysis,
which may not have pre-specified hypotheses. Statistical hypothesis testing is a key
technique of frequents inference.
Statistical hypothesis tests define a procedure that controls (fixes) the probability
of incorrectly deciding that a default position (null hypothesis) is incorrect based on how
likely it would be for a set of observations to occur if the null hypothesis were true. Note
that this probability of making an incorrect decision is not the probability that the null
hypothesis is true, nor whether any specific alternative hypothesis is true. This contrasts
with other possible techniques of decision theory in which the null and alternative
hypothesis are treated on a more equal basis. One naive Bayesian approach to hypothesis
testing is to base decisions on the posterior probability, but this fails when comparing
point and continuous hypotheses. Other approaches to decision making, such as
Bayesian decision theory, attempt to balance the consequences of incorrect decisions
across all possibilities, rather than concentrating on a single null hypothesis. A number
of other approaches to reaching a decision based on data are available via decision theory
and optimal decisions, some of which have desirable properties, yet hypothesis testing
is a dominant approach to data analysis in many fields of science. Extensions to the
theory of hypothesis testing include the study of the power of tests, which refers to the
probability of correctly rejecting the null hypothesis when a given state of nature exists.
Such considerations can be used for the purpose of sample size determination prior to
the collection of data.

Meaning of Hypothesis Testing

Hypothesis testing refers to the formal procedures used by statisticians to accept

or reject statistical hypotheses. It is an assumption about a population parameter. This
assumption may or may not be true.
The best way to determine whether a statistical hypothesis is true would be to
examine the entire population. Since that is often impractical, researchers typically
examine a random sample from the population. If sample data are not consistent with
the statistical hypothesis, the hypothesis is rejected.
In doing so, one has to take the help of certain assumptions or hypothetical values
about the characteristics of the population if some such information is available. Such
hypothesis about the population is termed as statistical hypothesis and the hypothesis
is tested on the basis of sample values. The procedure enables one to decide on a certain
hypothesis and test its significance. “A claim or hypothesis about the population
parameters is known as Null Hypothesis and is written as, H 0 .”

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 195

This hypothesis is then tested with available evidence and a decision is made whether Notes
to accept this hypothesis or reject it. If this hypothesis is rejected, then we accept the
alternate hypothesis. This hypothesis is written as H 1 .
For testing hypothesis or test of significance we use both parametric tests and
nonparametric or distribution free tests. Parametric tests assume within properties of the
population, from which we draw samples. Such assumptions may be about population
parameters, sample size etc. In case of non-parametric tests, we do not make such
assumptions. Here we assume only nominal or ordinal data.

Characteristics of Hypothesis

The important characteristics of Hypothesis are as follows:

(i) Hypothesis must be conceptually clear
The concepts used in the hypothesis should be clearly defined, operationally if
possible. Such definitions should be commonly accepted and easily communicable among
the research scholars.
(ii) Hypothesis should have empirical referents
The variables contained in the hypothesis should be empirical realities. In case these
are not empirical realities then it will not be possible to make the observations. Being
handicapped by the data collection, it may not be possible to test the hypothesis. Watch
for words like ought, should, bad.
(iii) Hypothesis must be specific
The hypothesis should not only be specific to a place and situation but also these
should be narrowed down with respect to its operation. Let there be no global use of
concepts whereby the researcher is using such a broad concept which may all inclusive
and may not be able to tell anything. For example somebody may try to propose the
relationship between urbanization and family size. Yes urbanization influences in declining
the size of families. But urbanization is such comprehensive variable which hide the
operation of so many other factor which emerge as part of the urbanization process. These
factors could be the rise in education levels, women's levels of education, women
empowerment, emergence of dual earner families, decline in patriarchy, accessibility to
health services, role of mass media, and could be more. Therefore the global use of the
word `urbanization' may not tell much. Hence it is suggested to that the hypothesis should
be specific.
(iv) Hypothesis should be related to available techniques of research
Hypothesis may have empirical reality; still we are looking for tools and techniques
that could be used for the collection of data. If the techniques are not there then the
researcher is handicapped. Therefore, either the techniques are already available or the
researcher is in a position to develop suitable techniques for the study.
(v) Hypothesis should be related to a body of theory
Hypothesis has to be supported by theoretical argumentation. For this purpose the
research may develop his/her theoretical framework which could help in the generation
of relevant hypothesis. For the development of a framework the researcher shall depend
on the existing body of knowledge. In such an effort a connection between the study in
hand and the existing body of knowledge can be established. That is how the study could
benefit from the existing knowledge and later on through testing the hypothesis could
contribute to the reservoir of knowledge.

Amity Directorate of Distance and Online Education

196 Quantitative Techniques in Management

Notes Source of Hypothesis

The various sources of hypotheses may be:

(i) Previous Study
Previous study is also a source of developing a concrete hypothesis. If a researcher
uses previous knowledge about a phenomenon for a particular place, then another
researcher followed his techniques and formulates his own.
(ii) Personal Experience
On the basis of his personal experience he uses his mind and suggests some points
for the eradication of a social problem through developing a good hypothesis. Greater the
researcher experience lead to higher degree of formation.
(iii) Imagination and Thinking
Creative thinking and imagination of a researcher sometimes help in formulating a
good hypothesis. Personal ideas and the thinking capabilities of a researcher would lead
to greater number of hypothesis formulation as well as control over the problem.
(iv) Observation
In consideration and undertaking a research problem, observation is necessary. The
collection of previous facts and current facts related to the problem lead to the formulation
of a good hypothesis.
(v) Scientific Theory
Theory is capable in explaining all the facts relating to the problem. Scientific theory
is a fertile source of hypothesis formulation. The theory which is used by a researcher
may satisfy the needs of making it, because theory explains the known facts.
(vi) Culture
Culture is the accumulation of ways of behaving and adoption in a particular place
and time. While formulating a hypothesis for a problem, culture should be studied. If we
want to study trends towards female education in a particular area, for this purpose we
will study, traditions, family system, Norms, Values, region and education system of that
area.

Formulation of Hypothesis

1. State the null hypothesis as well as the alternate hypothesis

For example, let us assume the population mean = 50 and set up the hypothesis
i = 50. This is called the null hypothesis and is denoted as;

Null hypothesis, H 0 : µ = 50

Alternative hypothesis H : µ = 50
1
or µ > 50
µ < 50
2. Establish a level of significance
The level of significance signifies the probability of committing Type 1 error á and
is generally taken as equal to 0.05. Sometimes, the value a is established as 0.01, but
it is at the discretion of the investigator to select its value, depending upon the sensitivity

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 197

of the study. To illustrate per cent level of significance indicates that a researcher is willing Notes
to take 5 per cent risk of rejecting the Null Hypothesis when it happens to be true.
3. Choosing a suitable test statistic
Now the researcher would choose amongst the various tests (i.e. z, t,  2 and f-
tests). Actually, for the purpose of rejecting or accepting the null hypothesis, a suitable
statistics called ‘test statistics’ is chosen. This means that H 0 is assumed to be really
true. Obviously due to sampling fluctuations, the observed value of the statistic based
on random sample will differ from the expected value. If the difference is large enough,
one suspects the validity of the assumption and rejects the null hypothesis ( H 0 ). On
the other hand, if the difference may be assumed due to sampling fluctuation, the null
hypothesis ( H 0 ) is accepted.
4. Defining the critical rejection regions and making calculations for test statistics
If we select the value of a = Level of significance = 0.05, and use the standard normal
distribution (z-test) as our test statistic for testing the population parameter u, then the
value of the difference between the assumption of null hypothesis (assumed value of the
population parameter) and the value obtained by the analysis of the sample results is not
expected to be more than than 1.96 s at a = 0.05.

Errors in Hypothesis

In statistical hypothesis testing, type I and type II errors are incorrect rejection of
a true null hypothesis or failure to reject a false null hypothesis, respectively. More simply
stated, a type I error is detecting an effect that is not present, while a type II error is
failing to detect an effect that is present. The terms "type I error" and "type II error" are
often used interchangeably with the general notion of false positives and false negatives
in binary classification, such as medical testing, but narrowly speaking refer specifically
to statistical hypothesis testing.
Two types of errors can result from a hypothesis test:
i) Type I error: A Type I error occurs when the researcher rejects a null hypothesis
when it is true. The probability of committing a Type I error is called the
significance level. This probability is also called alpha, and is often denoted by
.
ii) Type II error: A Type II error occurs when the researcher fails to reject a null
hypothesis that is false. The probability of committing a Type II error is called
Beta, and is often denoted by  . The probability of not committing a Type II
error is called the Power of the test.

4.11 Null Hypothesis and Alternative Hypothesis

Null Hypothesis

It is used for testing the hypothesis formulated by the researcher. Researchers treat
evidence that supports a hypothesis differently from the evidence that opposes it. They
give negative evidence more importance than to the positive one. It is because the negative
evidence tarnishes the hypothesis. It shows that the predictions made by the hypothesis
are wrong. The null hypothesis simply states that there is no relationship between the
variables or the relationship between the variables is "zero." That is how symbolically null

Amity Directorate of Distance and Online Education

198 Quantitative Techniques in Management

Notes
hypothesis is denoted as "H0". For example: H 0 = There is no relationship between the

level of job commitment and the level of efficiency. Or H 0 = The relationship between
level of job commitment and the level of efficiency is zero. Or The two variables are
independent of each other. It does not take into consideration the direction of association

(i.e. H 0 is non directional), which may be a second step in testing the hypothesis. First
we look whether or not there is an association then we go for the direction of association
and the strength of association. Experts recommend that we test our hypothesis indirectly
by testing the null hypothesis. In case we have any credibility in our hypothesis then the
research data should reject the null hypothesis. Rejection of the null hypothesis leads
to the acceptance of the alternative hypothesis.
Alternative Hypothesis
The alternative (to the null) hypothesis simply states that there is a relationship
between the variables under study. In our example it could be: there is a relationship
between the level of job commitment and the level of efficiency. Not only there is an
association between the two variables under study but also the relationship is perfect which
is indicated by the number "1". Thereby the alternative hypothesis is symbolically denoted

as " H 1 ". It can be written like this: H 1 : There is a relationship between the level of job
commitment of the officers and their level of efficiency.

4.12 t – test
A statistical examination of two population means. A two-sample t-test examines
whether two samples are different and is commonly used when the variances of two normal
distributions are unknown and when an experiment uses a small sample size.

X  
t
Formula: S
N
Where, is the sample mean, Ä is a specified value to be tested, s is the sample
standard deviation and n is the size of the sample. Look up the significance level of the
z-value in the standard normal table.
When the standard deviation of the sample is substituted for the standard deviation
of the population, the statistic does not have a normal distribution; it has what is called
the t-distribution. Because there is a different t-distribution for each sample size, it is not
practical to list a separate area of the curve table for each one. Instead, critical t-values
for common alpha levels (0.10, 0.05, 0.01, and so forth) are usually given in a single table
for a range of sample sizes. For very large samples, the t-distribution approximates the
standard normal (z) distribution. In practice, it is best to use t-distributions any time the
population standard deviation is not known.
Values in the t-table are not actually listed by sample size but by degrees of freedom
(df). The number of degrees of freedom for a problem involving the t-distribution for sample
size n is simply n – 1 for a one-sample mean problem.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 199

Uses of T Test Notes

Among the most frequently used t-tests are:
i) A one-sample location test of whether the mean of a normally distributed
population has a value specified in a null hypothesis.
ii) A two sample location test of the null hypothesis that the means of two normally
distributed populations are equal. All such tests are usually called Student’s
t-tests, though strictly speaking that name should only be used if the variances
of the two populations are also assumed to be equal; the form of the test used
when this assumption is dropped is sometimes called Welch’s t-test. These
tests are often referred to as “unpaired” or “independent samples” t-tests, as
they are typically applied when the statistical units underlying the two samples
being compared are non-overlapping.
iii) A test of the null hypothesis that the difference between two responses
measured on the same statistical unit has a mean value of zero. For example,
suppose we measure the size of a cancer patient’s tumor before and after a
treatment. If the treatment is effective, we expect the tumor size for many of
the patients to be smaller following the treatment. This is often referred to as
the “paired” or “repeated measures” t-test: A test of whether the slope of a
regression line differs significantly from 0.

Assumptions

Z
Most t-test statistics have the form T  , where Z and s are functions of the data.
S
Typically, Z is designed to be sensitive to the alternative hypothesis (i.e. its magnitude
tends to be larger when the alternative hypothesis is true), whereas s is a scaling
parameter that allows the distribution of T to be determined.

X
Z
As an example, in the one-sample t-test  , where
X is the sample mean
n
of the data, n is the sample size, and ó is the population standard deviation of the data; S
in the one-sample t-test is  /  , where  is the sample standard deviation.
The assumptions underlying a t-test are that:
i) Z follows a standard normal distribution under the null hypothesis

ii) p  2 follows a  2 distribution with p degrees of freedom under the null hypothesis,
where p is a positive constant
iii) Z and S are independent.

4.13 F – test
F-test is any statistical test in which the test statistic has an F-distribution under
the null hypothesis. It is most often used when comparing statistical models that have
been fitted to a data set, in order to identify the model that best fits the population from
which the data were sampled.

Amity Directorate of Distance and Online Education

200 Quantitative Techniques in Management

Notes Common examples of the use of F-tests are, for example, the study of the following
cases:
(i) The hypothesis that the means of a given set of normally distributed populations,
all having the same standard deviation, are equal. This is perhaps the best-
known F-test, and plays an important role in the analysis of variance (ANOVA).
(ii) The hypothesis that a proposed regression model fits the data well. See Lack-
of-fit sum of squares.
(iii) The hypothesis that a data set in a regression analysis follows the simpler of
two proposed linear models that are nested within each other.
The F-test is designed to test if two population variances are equal. It does this by
comparing the ratio of two variances. So, if the variances are equal, the ratio of the
variances will be 1.
If the null hypothesis is true, then the F test-statistic given above can be simplified
(dramatically). This ratio of sample variances will be test statistic used. If the null
hypothesis is false, then we will reject the null hypothesis that the ratio was equal to
1 and our assumption that they were equal.
There are several different F-tables. Each one has a different level of significance.
So, find the correct level of significance first, and then look up the numerator degrees
of freedom and the denominator degrees of freedom to find the critical value.
You will notice that all of the tables only give level of significance for right tail tests.
Because the F distribution is not symmetric and there are no negative values, you may
not simply take the opposite of the right critical value to find the left critical value. The
way to find a left critical value is to reverse the degrees of freedom, look up the right critical
value, and then take the reciprocal of this value. For example, the critical value with 0.05
on the left with 12 numerator and 15 denominator degrees of freedom is found of taking
the reciprocal of the critical value with 0.05 on the right with 15 numerator and 12
denominator degrees of freedom.

F Test Formula

F Test is a method to compare variance of two different set of values. F test is applied
on F distribution under null hypothesis. For calculating F test value, we first find the mean
of two given observations and then calculate their variance. F test value is expressed as
the ratio of variances of two observations. The comparison between the variances of two
sets of data can lead to many predictions. The formula for F test is given below:

variance 1  12
F value  
variance 2  22
Variance is given by the following formula:

 ( x  x )2
2
n 1

Where,

 2 = Variance
x = Values given in a set of data

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 201

X = Mean of the data Notes

n = Total number of values.

4.14 Non–Parametric Tests (chi – square test)

Chi-square Test
 2 test is a test that uses the chi-square statistic to test the fit between a theoretical
frequency distribution and a frequency distribution of observed data for which each
observation may fall into one of several classes.
Conditions of Chi-square (  2 ) Test
A chi-square  2 test can be used when the data satisfies four conditions:
i. There must be two observed sets of data or one observed set of data and one
expected set of data (generally, there are n-rows and c-columns of data).
ii. The two sets of data must be based on the same sample size.
iii. Each cell in the data contains the observed or expected count of five or large?
iv. The different cells in a row of column must have categorical variables (male,
female less than 25 years of age, 25 year of age, older than 40 years of age
etc.

Assumptions of Chi-square Test

The chi-squared test, when used with the standard approximation that a chi-squared
distribution is applicable, has the following assumptions:
i) Simple random sample: The sample data is a random sampling from a fixed
distribution or population where each member of the population has an equal
probability of selection. Variants of the test have been developed for complex
samples, such as where the data is weighted.
ii) Sample size (whole table): A sample with a sufficiently large size is assumed.
If a chi squared test is conducted on a sample with a smaller size, then the
chi squared test will yield an inaccurate inference. The researcher, by using
chi squared test on small samples, might end up committing a Type II error.
iii) Expected cell count: Adequate expected cell counts. Some require 5 or more,
and others require 10 or more. A common rule is 5 or more in all cells of a
2-by-2 table, and 5 or more in 80% of cells in larger tables, but no cells with
zero expected count. When this assumption is not met, Yates’s correction is
applied.
iv) Independence: The observations are always assumed to be independent of
each other. This means chi-squared cannot be used to test correlated data (like
matched pairs or panel data). In those cases you might want to turn to
McNamara’s test.

Application areas of Chi-square Test

The  2 distribution typically looks like a normal distribution, which is skewed to

the right with a long tail to the right. It is a continuous distribution with only positive values.
It has following applications:
i) To test whether the sample differences among various sample proportions are
significant or can they be attributed to chance.
Amity Directorate of Distance and Online Education
202 Quantitative Techniques in Management

Notes ii) To test the independence of two variables in a contingency table.

iii) To use it as a test of goodness of fit.

Degrees of Freedom (d.f)

The degree of freedom, abbreviated as d.f, denotes the extent of independence

(freedom) enjoyed by a given set of observed frequencies. Degrees of freedom are usually
denoted by the letter ‘v’ of the Greek alphabet.
Suppose, if we are given a set of ‘n’ observed frequencies which are subjected to
‘k’ independent constraints (restrictions). Then
Degrees of Freedom = No. of frequencies – No. of independent constraints v = n
– k
Formula of Chi-square text:

F bO g IJ
2

 G
 Ei
GH JK
2 i
x
Ei

Table value of x2 for d.f and 

X 2cal < X 2table , accept H0

4.15 Summary
Probability is the likelihood that an event will occur and is calculated by dividing the
number of favourable outcomes by the total number of possible outcomes.
Experiment refers to describe an act which can be repeated under some given
conditions. Random experiments are those experiments whose results depend on chance
such as tossing a coin, throwing a dice etc. The outcomes of random experiment are
events.
A probability distribution is a statistical function. It describes all the possible values
and possibilities that a random variable can take within a given range. This range will be
in between the minimum and maximum statistically possible values. But, it is where the
possible value is likely to be plotted on the probability distribution depends on a number
of factors. These factors include the distribution's mean, standard deviation, skewness
and kurtosis.
The Poisson distribution is used to describe discrete quantitative data such as counts
in which the populations size n is large, the probability of an individual event  is small,
but the expected number of events, n  is moderate say five or more.
Hypergeometric distribution is a probability distribution that’s very similar to the
binomial distribution. In fact, the binomial distribution is a very good approximation of the
hypergeometric distribution as long as you are sampling 5% or less of the population.
Non-probability samples that are unrestricted are called convenient sampling.
Convenience sampling refers to the collection of information from members of population
who are conveniently available to provide it. Researchers or field workers have the freedom
to choose as samples whomever they find, thus it is named as convenience.
Judgment sampling is a non-probability sampling method in which participants are
selected according to an experienced individual’s belief that they will meet the

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 203

requirements of the study. The researcher selects sample members who conform to some Notes
criterion.
Snowball sampling is a non-probability sampling method in which a set of
respondents are chosen who help the researcher to identify additional respondents to be
included in the study. This method of sampling is also called as referral sampling because
one respondent refers other potential respondents. This method involves probability and
non-probability methods.
Hypothesis test is a method of making decisions using data from a scientific study.
In statistics, a result is called statistically significant if it has been predicted as unlikely
to have occurred by chance alone, according to a pre-determined threshold probability,
the significance level.
F-test is any statistical test in which the test statistic has an F-distribution under
the null hypothesis. It is most often used when comparing statistical models that have
been fitted to a data set, in order to identify the model that best fits the population from
which the data were sampled.
Chi-square test is a test that uses the chi-square statistic to test the fit between
a theoretical frequency distribution and a frequency distribution of observed data for which
each observation may fall into one of several classes.

4.16 Check Your Progress

I. Fill in the Blanks

1. Measure of chance is known as ____________
2. Probability distribution of discrete random variable is classified as ___________
3. Probability distribution having shape of bell and in which values of mean lies
in center of probability distribution is classified as _________
4. Discrete probability distribution in which outcome is very small with a very small
period of time is classified as ________
5. Nature of the binomial random variable X is ___________
6. Formula of calculating mean for hyper geometric probability distribution is
____________
7. Branch of statistics in which data is collected according to ordinal scale or
nominal scale is classified as__________________

II. True or False

1. Outcomes of an experiment are classified as event.
2. Value which is used to measure distance between mean and random variable
x in terms of standard deviation is called Probability of x.
3. If mean of binomial probability distribution is 25 then mean of Poisson probability
distribution is 25.
4. Probability which explains x is equal to or less than particular value is classified
as Cumulative probability.
5. Z test is applied when samples are less than 30.

Amity Directorate of Distance and Online Education

204 Quantitative Techniques in Management

Notes III. Multiple Choice Questions

1. Probability of second event in situation if first event has been occurred is
classified as
[a] Series probability
[b] Conditional probability
[c] Joint probability
[d] Dependent probability
2. Probability of failure in binomial distribution is denoted by
[a] p = q +1
[b] p = q - 1
[c] q = 1 + p
[d] q = 1 - p
3. Probability which is based on self-beliefs of persons involved in experiment is
classified as
[a] Subjective approach
[b] Objective approach
[c] Intuitive approach
[d] Sample approach
4. In probability theories, events which can never occur together are classified as
[a] Collectively exclusive events
[b] Mutually exhaustive events
[c] Mutually exclusive events
[d] Collectively exhaustive events
5. Conditional probability of two events Y and Z written as P(Z|Y) = P(Y and Z)
/ Pa) shows that events are
[a] Statistically dependent events
[b] Descriptive unaffected events
[c] Statistically independent events
[d] Statistically unaffected events
6. Value which is used to measure distance between mean and random variable
x in terms of standard deviation is called
[a] Z-value
[b] Variance
[c] Probability of x
[d] Density function of x
7. Variance of random variable x of gamma distribution can be calculated as
[a] Var(x) = n + 2 / µsup2
[b] Var(x) = n / µsup2
[c] Var (x) = n * 2 / µsup2
[d] Var(x) = n - 2 / µsup3
8. Type of distribution which is useful when occurrences of events are constant
is classified as
[a] Open frequency distribution
Amity Directorate of Distance and Online Education
Probability and Testing of Hypothesis 205

[b] Class frequency distribution Notes

[c] Rectangular distribution
[d] Square distribution
9. __________ test is applied when samples are less than 30.
[a] T
[b] Z
[c] Rank
[d] None of these
10. Under non-random sampling method, samples are selected on the basis of
[a] Stages
[b] Strategy
[c] Originality
[d] Convenience

4.17 Questions and Exercises

I. Short Answer Questions

1. What is probability?
2. Define experiment.
3. Define event.
4. What do you mean by probability distributions?
5. What is continuous probability distributions?
6. What is binomial distribution?
7. Give the meaning of poisson distribution.
8. What do you mean by hypergeometric distribution?
9. What is normal distribution?
10. What is exponential distribution?
11. What is sampling?
12. What is stratified random sampling?
13. What is convenience sampling?
14. What is systematic sampling?
15. What is snowball sampling?
16. What is quota sampling?
17. Define stage sampling or multi stage sampling.
18. What is judgement sampling?
19. What is cluster sampling?
20. What is sequential sampling?
21. What is On-the-spot-sampling?
22. What is Hypothesis Testing?
23. What is Null Hypothesis?
24. What do you mean by t-test?

Amity Directorate of Distance and Online Education

206 Quantitative Techniques in Management

Notes 25. What is F-test?

26. What is Chi-Square Test?

II. Extended Answer Questions

1. Write short note on probability.
2. Explain the basic structure of probability.
3. Explain the properties of normal distribution.
4. What are the merits and demerits of sampling?
5. Explain the various steps involved in sampling process.
6. Define sampling. Explain the need and significance of sampling.
7. Explain various technique of sampling.
8. What are characteristics of hypothesis?
9. Explain the types of hypothesis.
10. Discuss the source of hypothesis.
11. Explain the formulation of hypothesis.
12. Explain the null hypothesis and alternative hypothesis.
13. What are the uses of t-test?
14. Write short note on F-test.
15. Explain the assumptions of Chi-square Test.
16. Discuss the application areas of Chi-square Test.

4.18 Key Terms

 Probability: Probability of a given event is an expression of likelihood or chance
of occurrence of an event. A probability is a number which rages from zero to
one.
 Continuous Probability Distributions: Continuous random variables are those
that take on any value including fractions and decimals. Continuous random
variables give rise to continuous probability distributions. Continuous is the
opposite of discrete.
 Sample: A sample is that part of the universe which the select for the purpose
of investigation. A sample exhibits the characteristics of the universe. The word
sample literally means small universe.
 Sampling: Sampling is defined as the selection of some part of an aggregate
or totality on the basis of which a judgment or inference about the aggregate
or totality is made. Sampling is the process of learning about the population
on the basis of a sample drawn from it.
 Stratified random sampling: Stratified random sampling requires the
separation of defined target population into different groups called strata and
the selection of sample from each stratum.
 Cluster sampling: Cluster sampling is a probability sampling method in which
the sampling units are divided into mutually exclusive and collectively exhaustive
subpopulation called clusters.
 Hypothesis testing: Hypothesis testing refers to the formal procedures used
by statisticians to accept or reject statistical hypotheses. It is an assumption
about a population parameter. This assumption may or may not be true.

Amity Directorate of Distance and Online Education

Probability and Testing of Hypothesis 207

4.19 Check Your Progress: Answers Notes

I. Fill in the Blanks
1. Probability 2. Probability mass function
3. Normal distribution 4. Poisson distribution
5. Discrete 6. n (m / n)
7. Non-parametric statistics
II. True or False
1. True 2. False
3. True 4. True
5. False
III. Multiple Choice Questions
1. (b) 2. (d)
3. (a) 4. (c)
5. (a) 6. (a)
7. (d) 8. (c)
9. (a) 10. (d)

4.20 Case Study

The tea stall near the railway station at Yesvantpura has been having average sale
of 500 tea cups per day. Because of the development of bus stand nearby, it expects
its sales to increase. During the first 12 days after the start of the bus stand, the daily
sales were as under:
550, 570, 490, 615, 505, 580, 570, 460, 600, 580, 530, 526. On the basis of sample
information, can one conclude that the tea stall’s sales have increased? (take  = 0.05).

4.21 Further Readings

1. Richard I. Levin, David S. Rubin, Sanjay Rastogi Masood Husain Siddiqui,
Statistics for Management, Pearson Education, 7th Edition, 2016.
2. Prem.S.Mann, Introductory Statistics, 7th Edition, Wiley India, 2016.
3. Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani, An Introduction
to Statistical Learning with Applications in R, Springer, 2016.
4. Aczel A.D. and Sounderpandian J., “Complete Business Statistics”, 6th edition,
Tata McGraw Hill Publishing Company Ltd., New Delhi, 2012.
5. Anderson D.R., Sweeney D.J. and Williams T.A., Statistics for business and
economics, 11th edition, Thomson (South – Western) Asia, Singapore, 2012.

4.22 Bibliography
1. Srivastava V. K. etal – Quantitative Techniques for Managerial Decision Making,
Wiley Eastern Ltd
2. Richard, I.Levin and Charles A.Kirkpatrick – Quantitative Approaches to
Management, McGraw Hill, Kogakusha Ltd.
3. Prem.S.Mann, Introductory Statistics, 7th Edition, Wiley India, 2016.

Amity Directorate of Distance and Online Education

208 Quantitative Techniques in Management

Notes 4. Aczel A.D. and Sounderpandian J., “Complete Business Statistics”, 6th edition,
Tata McGraw Hill Publishing Company Ltd., New Delhi, 2012.
5. Anderson D.R., Sweeney D.J. and Williams T.A., Statistics for business and
economics, 11th edition, Thomson (South – Western) Asia, Singapore, 2012.
6. Budnik, Frank S Dennis Mcleaavey, Richard Mojena – Principles of Operation
Research - AIT BS New Delhi.
7. Sharma J K – Operation Research- theory and applications-Mc Millan,New Delhi
8. Kalavathy S. – Operation Research – Vikas Pub Co
9. Gould F J – Introduction to Management Science – Englewood Cliffs N J Prentice
Hall.
10. Naray J K, Operation Research, theory and applications – Mc Millan, New Dehi.
11. Taha Hamdy, Operations Research, Prentice Hall of India
12. Tulasian: Quantitative Techniques: Pearson Ed.
13. Vohr.N.D. Quantitative Techniques in Management, TMH.
11. Stevenson W.D, Introduction to Management Science, TMH.


Amity Directorate of Distance and Online Education