Unit 2

Correlation
2 variables are said to be corelated if the change in 1 variable results in a corresponding change in
the other variable, i.e., when 2 variables move together, we say that they are corelated.

For example, when the price of a commodity races the supply for that commodity also races.

On the other hand, if price falls the supply also falls.

Positive and negative correlation

When the values of 2 variables move in the same direction, correlation is said to be positive.

i.e., an increase in the value of 1 variable results into an increase in the value of the other variable, or
if a decrease in the value of 1 variable results into a decrease in the value of the other variable.

for example, price and supply are positively corelated.

X Y
1 2
2 4
3 6
4 8
The value of 2 variable moves in opposite direction so that an increase in the value of 1 variable
results into a decrease in the value of the other variable, the correlation is said to be negative.

For example, price and demand are negatively correlated.

X Y
1 4
2 3
3 2
4 1
Linear and nonlinear correlation
When the amount of change in 1 variable leads to a constant ratio of change in the other variable,
correlation is said to be linear.

When the amount of change in 1 variable is not in constant ratio to the change in the other variable,
the correlation is said to be nonlinear.

X Y
1 1
2 2
3 3
4 4
Simple, multiple and partial correlation
In the study of relationship between variables, if there are only 2 variables, the correlation is said to
be simple.

e.g., price and demand.

In the study of multiple correlation, we measure the degree of association between 1 variable on 1
side and all the other variables together on the other side.

Thuse, the relationship between rain and temperature together is multiple correlation.

In partial correlation, we study the relationship of 1 variable with 1 of the other variable assuming
that, the other variables remain constant.

e.g., There are 3 variables, yield, rain and temperature. The relationship between yield and rain
(assuming temperature is constant) is the partial correlation.

Methods for studying correlation

Coefficient of correlation
Coefficient of correlation is an algebraic method of measuring correlation. It is a numerical value.

Coefficient of correlation is a pure number lying between -1 and 1.

When the correlation is negative, it lies between -1 and 0

when the correlation is positive, it lies between 0 and 1

When the coefficient of correlation is 0, it indicates that there is no correlation between the
variables.

When the coefficient of correlation is 1, there is positive perfect correlation

When the coefficient of correlation is -1, there is negative perfect correlation.

Coefficient of correlation can be computed by applying the method given below:

Karl Pearson’s correlation method

Spear man’s method

Concurrent deviation method

Karl Pearson’s correlation method

The formula for computing coefficient of correlation is

n Σ ( xy ) −Σ ( x ) Σ ( y )
r=
√ n Σ ( x )−( Σ x ) √ n Σ ( y ) −( Σ y )
2 2 2 2

n = number of pairs of observations.

Q: Calculate coefficient of correlation

x 2 3 4 5 6 7 8
y 4 5 6 12 9 5 4
xy 8 15 24 60 54 35 32
x
2
4 9 16 25 36 49 64
y
2
16 25 36 144 81 25 16
Σ x = 35

Σ y = 45

Σ (xy) = 228
 2
Σ x =203
2
Σ y =343

( Σ x )2=1225

( Σ y )2=2025
N=7

n Σ ( xy )−Σ ( x ) Σ ( y )
r=
√ n Σ ( x ) −( Σ x ) √n Σ ( y )− ( Σ y )
2 2 2 2

(7 × 228 )−(35 × 45)

¿
√(7 ×203)−1225 √(7 × 343)−2025
21
¿
√196 √ 376
21
¿
14 ×19.39
21
¿
271.47
¿ 0.077
Q: Compute the coefficient of correlation between price and demand

Price (x), 7, 8, 9, 6, 5

Demand (y), 8, 6, 7, 9, 10

Answer = -0.9

x 7 8 9 6 5
y 8 6 7 9 10
xy 56 48 63 54 50
x
2
49 64 81 36 25
y
2
64 36 49 81 100
n=5

Σ x = 35

Σ y = 40

Σ (xy) = 271
2
Σ x =255
2
Σ y =330

( Σ x )2=1225

( Σ y )2=1600
 n Σ ( xy )−Σ ( x ) Σ ( y )
r=
√ n Σ ( x ) −( Σ x ) √n Σ ( y )− ( Σ y )
2 2 2 2

( 5 ×271 )− (35 × 40 )
¿
√( 5 × 255 )−1225 √ ( 5× 330 )−1600
−45
¿
√50 √ 50
−45
¿
7.071× 7.071
−45
¿
50
¿−0.9
Properties of correlation coefficient
Correlation coefficient has a well-defined formula.

It is a pure number and it is independent of the units of measurement.

It lies between -1 and 1

Coefficient of correlation between x and y is same as that between y and x.

Coefficient of correlation doesn’t change with reference to change of origin or change of scale.

Probable error and standard error

Probable error of the coefficient of correlation is a statistical measure which measures reliability and
dependability of the value of coefficient of correlation.

The formula for finding probable error is

0.6745 ( 1−r 2 )
Probable error ¿
√n
r = coefficient correlation

n = number of pares of observations

1−r 2
standard error (se) ¿
√n
q: if r = 0.6 n = 64 find probable error and se

0.6745 ( 1−r 2 )
probable error ¿
√n
0.6745 ( 1−0.6 2 )
¿
√ 64
Q: if r = 0.89 probable error = 0.023 find n

Q: calculate Karl Pierson’s coefficient of correlation between x and y from the following data
N= 10

Σ x = 35

Σ y = 28

Σ x^2 = 203

Σ y^2 = 140

Σ xy = 168

n Σ ( xy ) −Σ ( x ) Σ ( y )
r=
√ n Σ ( x )−( Σ x ) √ n Σ ( y ) −( Σ y )
2 2 2 2

( 10 ×168 )−( 28 ×35 )

=
√10 × Σ ( x )−( Σ x ) √ 10 Σ ( y ) −( Σ y )
2 2 2 2

700
¿
√805 √616
700
¿
704.18747503772033505607095349378
¿ 0.99
Spearman’s method (rank correlation)
2
6ΣD
According to spearman’s method, the formula for rank correlation coefficient is 1−
n ( n 2−1 )
Where d is difference between ranks, n is number of observations

Q: the ranking of 10 individuals of the start and at the finish of a course of a training is as follows:

Individuals, a, b, c, d, e, f, g, h, I, j

Rank before, 1,6,3,9,5,2,7,10,8,4

Rank after, 6,8,3,2,7,10,5,9,4,1

D,5,2,0,7,2,82,1,4,3
2
d ,5,2,0,7,2,82,1,4,3
n=10
2
Σ d =106
2
6ΣD
r =1−
n ( n2−1 )
6 ×106
r =1−
10 ( 99 )
6 ×106
r =1−
10 ( 99 )
 r =0.357
q: 10 competitors in a beauty contest are ranked by 3 judges in the following order:

judge 1,1,6,5,10,3,2,4,9,7,8

judge 2,3,5,8,4,7,10,2,1,6,9

judge 3,6,4,9,8,1,2,3,10,5,7

use the correlation coefficient to discuss which pare of judges have nearest approach to common
taste in beauty.

Pare 1:

judge 1,1,6,5,10,3,2,4,9,7,8

judge 2,3,5,8,4,7,10,2,1,6,9

d,2,1,3,6,4,8,2,8,1,1
2
d ,4,1,9,36,16,64,4,64,1,1
2
Σ d =200
n=10

6 ( 200 )
Rank correlation ¿ 1−
10 ( 100−1 )
6 ( 200 )
¿ 1−
10 ( 100−1 )
¿−0.212
Pare 2:

judge 2,3,5,8,4,7,10,2,1,6,9

judge 3,6,4,9,8,1,2,3,10,5,7

d,3,1,1,4,6,8,1,9,1,2
2
D ,9,1,1,16,36,64,1,81,1,4
2
Σ d =214
n=10

6 ( 214 )
Rank correlation ¿ 1−
10 ( 100−1 )
6 ( 214 )
¿ 1−
10 ( 100−1 )
¿−0.296
Pare 3:
judge 1,1,6,5,10,3,2,4,9,7,8

judge 3,6,4,9,8,1,2,3,10,5,7

d,5,2,4,2,2,0,1,1,2,1
2
d ,25,4,16,4,4,0,1,1,4,1
2
Σ d =60
n=10

6 ( 60 )
Rank correlation ¿ 1−
10 ( 100−1 )
¿ 0.636
Q:

X,6,8,12,15,18,20,24,28,31

Y,10,12,15,15,18,25,22,26,28

Q:

X,50,60,55,65,75,70m75,80,90,80

Y, 10,14,15,11,12,15,16,20,18,19

Q: find the rank correlation coefficient between poverty and overcrowding from the following table

X, 17,13,15,16,6,11,14,9,7,12

Rank (x), 1, 5, 3, 2, 10 7, 4, 8, 9, 6

Y, 36,46,35,24,12,18,27,22,2,8

Rank (y), 2,1,3,5,8,7,4,6,10,9

D,1,4,0,3,2,0,0,2,1,3
2
d ,1,16,0,9,4,0,0,4,1,9
2
Σ d =44
6 ( 44 )
Rank correlation ¿ 1−
10 ( 100−1 )

Repeated rank (tie in rank)

Rank correlation coefficient = 1−¿

Where, m = number of times each value repeats

Q: Obtain the rank correlation coefficient of the following data

X,68,64,75,50,64,80,75,40,55,64

Y,62,58,68,45,81,16,68,48,50,70
Rank (x), in x, 64 accurse 3 times so m = 3

Q: the ranks of 11 students in 2 tests are given below∷

Test 1 ,80,45,55,58,55,60,45,68,70,45,85

Test 2,82,56,50,43,56,62,64,65,70,64,90

Calculate the coefficient of correlation by method of rank difference.

Scatter diagram
This is a graphical method of studying correlation between 2 variables

Scatter diagram is a visual aid to show the presents or absence of correlation between 2 variables

One of the variable is shown x axis and other on the y axis

Each pair of values plotted on the graph by means of dot mark.

Regression analysis
It is the estimation of or the prediction of the unknown value of 1 variable from the known-value at
the other variable

It is a statistical device used to study the relationship between 2 or more variables that are related

Dependent and independent variables

The variable whose value influenced or is to be predicted is called dependent variable.

The variable which influences the values or is used for prediction is called independent variables.

Simple and multiple regression

When there are only 2 variables the regression equation obtained is called simple regression
equation.

In multiple regression analysis there are more than 2 variables and we try to find out the effect of 2
or more independent variable on 1 depend variable.

Linear and non-linear regression

If the regression curve is a straight line, we say that it is linear regression between variable under the
study

If the regression curve is not a straight line, we say that it is nonlinear regression between variable
under the study

Properties of Regression coefficient

The sign of both the regression coefficients is the same

( byx ) × ( bxy ) =r 2
byx and bxy will have the same sign as r

( bxy ) =r ( σxσy )
Both the regression coefficients will not be greater than 1. i.e., one of them can be greater than 1 or
both can be less than 1.
Properties of regression line
2 lines intercept at ( x ' , y ' )

When r = 1 2 lines coincides

When r = 0 2 lines are mutually perpendicular

Regression equation of y on x
y− y ' =( byx ) ( x −x' )

( byx )=
( n Σ xy−(Σ x × Σy)
2
n Σ x −( Σ x )
2
)
' Σx
x=
n
' Σy
y=
n
Regression equation of x on y
x−x ' =( bxy ) ( y− y ' )

( bxy ) =
( n Σ xy −(Σ x . Σy)
2
n Σ y −(Σ y )
2
)
Where, byx and bxy are known as regression coefficients

Q: from the following data find regression of y on x

x,2,3,4,5,6

y,3,5,4,8,9

Q: find the regression equation of x on y

x 5 6 7 3 2 23
y 4 5 8 2 1 23
2
y

( bxy ) =
( n Σ ( xy ) −(Σ x × Σy)
n Σ y 2− ( Σ y )
2 )
Q: the following data of age of husband-and-wife form 2 regression equations and calculate the
husband’s age when the wife’s age 16 also find the age of wife when husband’s age 40

X,36,23,27,28,28,29,30,31,33,35

Y,29,18,20,22,27,21,29,27,29,28

When wife’s age y = 16

Relation between correlation coefficient (r) regression coefficient (byx, and bxy)

Byx=r ¿
 Bxy=r ¿
2
r = ( byx )( bxy )

r =√ ( byx )( bxy )
Q: find r if byx = -.2 bxy = -.7

r =√ ( byx )( bxy )
= -0.374

Q: Byx = .83, σ x = 20, σ y = 12, find r

Byx = r ( σyσx )
q: find byx if 2x+4y-5 = 0 is the equation of y on x

Given 2x+4y-5=0,

Q: given x’ = 36 y’ = 85 σ x =11 σ y = 8 r = .66

Find the 2 regression equations, estimate the value of x when y = 75

Method of least square

This is a method to find the unknown coefficient in a curve that servs the best approximation to the
curve y = f(x)

The principle of least square says that the sum of the squares of the error between the observed
values and the corresponding estimated value should be the least.

Fitting of a straight line

The equation of a straight line is y = a+(bx)

Where, a and b are parameters to be determined.

Normal equations are

σ y = ( na )+ bσ ( x )

σ (xy) = ¿

solving these 2 normal equations we get a and b

q: fit a straight line to the following data. Estimate the value of y when x = 3.5

x,1,2,3,4,5

y,14,13,4,5,2

xy,14,26,12,20,10
2
x ,1,4,9,16,25
σ x = 15

σ y = 38
σ xy = 82

σ x^2 = 55

the equation of the strait line is y=a+(bx)

the normal equations are

38 = ( 5 a )+15 b

82 = ( 15 a )+ (55 b )

Equation 1, ( 5 a )+ ( 15 b )=38

Equation 2, ( 15 a )+ (55 b )=82

Equation 3, equation 1 × 3 → ( 15 a )+ ( 45 b ) =114

Equation 3 – equation 2

Equation 3, equation 1 × 3 → ( 15 a )+ ( 45 b ) =114

Equation 2, ( 15 a )+ (55 b )=82

-10b = 32

32
b=
−10
b=−3.2
Substitute b = -3.2 on equation 1

( 5 a )+ ( 15 (−3.2 )) =38

( 5 a )=38+48
( 5 a )=86
86
a=
5
a=17.2
Apply a and b to y = a+bx

y=17.2 x−3.2
Applying x as 3.5 to the equation

y=17.2 ( 3.5 ) −3.2

¿ 57
So, y is 57 when x is 3.5

Q: fit a straight-line y = a+bx to the following data

X,1,2,3,4,6,8
Y,2.4,3,3.6,4,5,6
2
x ,1,4,9,16,36,64
Xy,2.4,6,10.8,16,30,48

σ x = 24

σ y = 24

σ (xy) = 113.2
2
σx =130
Normal equations are:

24=( 6 a )+ 24 b
113.2=( 24 a )+ ( 130 b )
Equation 1, ( 6 a )+ 24 b=24

Equation 2, ( 24 a ) + ( 130 b )=113.2

Equation 3, Multiplied equation 1 by 4

Equation 3, ( 24 a ) +96 b=96

Subtracting equation 3 – equation 2

Equation 3, ( 24 a ) +96 b=96

Equation 2, ( 24 a ) + ( 130 b )=113.2

34b= 17.2

17.2
b=
34
b=0.505
Substitute b = 0.505 on equation 1

( 6 a )+ 24 ( 0.505 ) =24
( 6 a )+ 12.12=24
6 a=(24−12.12)
6 a=11.88
11.88
a=
6
a=1.98
Substituting a and b to the equation y

y=1.98+ 0.505 x
Notes
when the equation of strait line is y = ax+b

the normal equations are:

σy= ( nb ) +a Σ x
2
σ (xy)=aσ x + bσ x
q: fit a straight line by the method of least square for the following data

x,0,1,2,3,4

y,1,1.8,3.3,4.5,6.3

x^2, 0,1,4,9,16

xy,0,1.8,6.6,13.5,25.2

σ x = 10

σ y = 16.9

σ x^2 = 30

σ (xy) = 47.1

normal equations

16.9= ( 5b ) +a 10
47.1=a 30+b 10
Equation 1, 10 a+ ( 5 b )=16.9

Equation 2, 30 a+10 b=47.1

Equation 3, Equation 1 × 3 → 30a+15b=50.7

Equation 3 – equation 2

Equation 3, 30a+15b=50.7

Equation 2, 30 a+10 b=47.1

5b=3.6

3.6
b= =0.72
5
Substitute b = 0.72 in equation 1

Equation 1, 10 a+ ( 5 ( 0.72 ) )=16.9

10 a+ ( 3.6 )=16.9
10 a=16.9−( 3.6 )
10 a=13.3
 13.3
a=
10
a=1.33
y=ax+ b
y=1.33 x +0.72
Probability
The probability of a given event may be defined as the numerical value given to the likelihood of the
occurrence of that event.

It is a number lying between 0 and 1

0 is for an event which can’t occur and 1 for an event certain to occur (sure)

e.g., when we toss a coin the event of getting head is uncertain so the probability is neither 0 nor 1
but between the 2.

1
Therefore, the probability for getting head =
2
Random experiment
An experiment that has 2 or more outcomes which various unpredictable manor from trial to trial
which is conducted under uniform conditions is called a random experiment.

In a random experiment, all the possible outcomes are known in advance.

But, nun of the outcomes can be predicted with certainty.

Essential features of random experiment are as follows:

It has more than one outcome.

The outcomes are unpredictable

The experiment is repeatable

Sample point, simple event, or elementary outcomes

Every indecomposable outcome of a random experiment is called sample point of the random
experiment

e.g., when a coin is tossed the probability of getting head is a sample point

sample space
the sample space of a random experiment is the set containing all the sample points of that random
experiment

e.g., when a coin is tossed, the sample space is { head , tail }

when 2 coins are tossed the sample space is { ( head , tail ) , ( head , head ) , ( tail , head ) , ( tail , tail ) }

q: a box contains 10 tickets, each numbered from 1 to 10. A ticket is drawn. What is the sample
space?

Sample space = { 1 , 2, 3 , 4 ,5 , 6 , 7 , 8 , 9 ,10 }

Q: from lot containing good and bad item 3 items are chosen prepare the sample space

Sample space = { ( g , g , g ) , ( g , g ,b ) , ( g , b , g ) , ( g , b , b ) , ( b , g , g ) , ( b , g , b ) , ( b , b , g ) , ( b ,b ,b ) }

Q: 2 dice are thrown find the probability that:

1. Both the dice show the same number

2. 1 die shows 5
3. First die shows 5
4. The total of the numbers on the dice is 8
5. The total of the numbers on the dice is greater than 8
6. A sum of 10

sample space =
{ ( 1 , 1 ) , ( 1 , 2 ) , ( 1 , 3 ) , ( 1 , 4 j ) , ( 1 , 5 ) , ( 1 , 6 ) , ( 2 ,1 ) , ( 2 ,2 ) , ( 2 ,3 ) , ( 2 , 4 ) , ( 2 , 5 ) , ( 2 , 6 ) , ( 3 ,1 ) , … , ( 6 , 6 ) }
Total number of cases = 36

6
P ( both dice shows same number )
36
10
p ( 1 die shows 5 )=
36
6
p ( first die shows 5 )=
36
5
p ( thetotal of the numbers of the dice are 8 )=
36
10
P (The total of the numbers on the dice is greater than 8) =
36
3
P (a sum of 10) =
36
Q: a ball is drawn from a bag containing 4 whites 6 black and 5 green balls. Find the
probability that a ball is drawne:
1. White
2. Green
3. Black
4. Not green
5. Green or white

4
p ( drawing a white ball )=
15
5
p ( drawing a green ball ) =
15
6
P (drawing a black ball) =
15
10
p ( not green )= 15
❑
 9
p ( green∨white )=
15
Q: a card is drawn from pack of cards. What is the probability that it is:

1. A black card
2. A king
3. A queen
4. A spade
5. A spade king
6. A king or a queen

Sample space

sample space={ Ace of Spades(♠), 2 of Spades , 3 of Spades , 4 of Spades ,5 of Spades , 6 of Spades , 7 of Spades

Total number of cases = 52

26
p ( drawing a black card )=
52
4
P (drawing a king) ¿
52
4
p ( drawing a queen )=
52
13
p ( drawing a spade )=
52
1
p ( drawing a spade king )=
52
8
p ( drawing a king∨queen )=
52
Q: what is the probability that a leap year selected at random will contain 53 sundays

A leap year has 366 days

So, it has 52 weeks and 2 days

52 weeks have 52 Sundays. So, a leap year will have 53 Sundays only if the last 2 days of the year is a
Sunday.

There are 7 possibilities and favorable cases Sunday=2

2
Therefore, p (getting 53 Sundays) =
7
Q: what is the probability of getting 3 white balls in a draw of 3 balls from a box containing 5 whites
and 4 blacks

Favorable number of cases = 5c3

Total number of cases = 9c3

5 c 3 10
Therefore p ( getting 3 white balls )= =
9 c 3 84
Q: a bag contains 7 white and 9 black balls. 3 balls are drawn together. What is the probability that:

1. All are blacks

2. All are whites
3. 1 white and 2 blacks
4. 2 whites and 1 black

9c3 7c3
p ( all are blacks )= p ( all are whites )=
16 c 3 16 c 3

( 7 c 1) ( 9 c 2)
p ( 1 white∧2 blacks ) =
16 c 3
( 7 c 2 )( 9 c 1 )
p ( 2 whites∧1 black )=
16 c 3