Chapter 1 | Data & Statistics

↪ What is statistics?
-​ The art and science of COLLECTING, ANALYZING, PRESENTING and INTERPRETING data

** stats makes SENSE of numbers + informations **

> data = facts & figures

> data set = facts & figures collected for a study

a. Elements

-​ The entities on which data are collected from

-​ From WHO are we collecting data

-​ From WHAT are we collecting data

b. Variables

-​ A characteristic of interest for the elements

-​ WHAT information are we collecting

c. observations

-​ The set of measurements obtained for a particular element

-​ EVERYTHING collected from a single element

> an entire row of data

** the number of observations will be EQUIVALENT to the number of elements **

↪ Scale of Measurement
-​ This determines the amount of information contained in the data, indicating the most appropriate
form of summarization + analysis

-​ Involves nominal, ordinal, ratio + interval scales ( N.O.I.R )

Quantitative Data Qualitative / Categorical Data
- Numerical data ( CAN calculate ) - non numerical data ( CANNOT calculate )

⬇ ⬇ ⬇ ⬇

Continuous data Discrete data Nominal scale Ordinal scale

- this deals with - this deals with - there is NO order, just - has a meaningful rank
FRACTIONS & INTEGERS & WHOLE shelving items
DECIMALS NUMBERS E.g. excellent, average,
( measurable ) ( can be counted ) poor

⬇ ↘ ↙ ⬇

Ratio scale Interval scale

- has a positive or - has NO true zero
negative outcome
- has a TRUE ZERO E.g. 0*C doesn’t mean
there in no temperature,
E.g. yes / no , a / b it is only VERY COLD
 -​ summarized data can be displayed using the ‘pivot table’

↪ Collecting Data

a. Cross-sectional data

-​ Data collected at the SAME time or same POINT in time

-​ Data collected once

b. Time series data

-​ Data collected over several periods of time

> where is data collected?

1.​ Population
-​ A set of all elements of interest in a particular study

-​ A Large group of elements

2.​ Sample
-​ A subset of the population

-​ A smaller group of the population

> how data can be collected?

Descriptive Statistics / Analytics Inferential Statistics / Analytics

- take data to DESCRIBE / ILLUSTRATE it - take data to PREDICT / CONCLUDE / ESTIMATE
it

⬇ ⬇

1.​ Queries 1.​ Census

-​ A survey to collect data on an
2.​ Reports ENTIRE POPULATION
3.​ Data visualization
2.​ Sample survey
E.g. pie charts, bar graphs, frequency tables -​ A survey to collect data on a
SAMPLE

↪ Data Sets

a. Big data

-​ Data that CANNOT be managed, processed or analyzed with available software in an amount of
time

-​ Large amounts / variety of data

b. Data mining

-​ Process of using STATS + COMPUTER SCIENCE to obtain useful information from LARGE
databases
 Chapter 2 | Tabular & Graphical Displays
↪ Summarizing Data

a. Frequency Distribution

-​ A tabular summary of data, showing the NUMBER of observations in each NON-OVERLAPPING

categories / classes

E.g. different types of SOFT DRINKS, different types of SOUR SWEETS

b. Relative Frequency

-​ A tabular summary of data, showing the fraction / PROPORTION of observations in each

NON-OVERLAPPING classes

> calculation = class frequency ➗ number of observations ( ‘n’ )

c. Percentage Frequency

-​ A tabular summary of data showing the PERCENTAGE of observations in each of several

NON-OVERLAPPING classes

> calculation = ( class frequency ➗ number of observations ) ✖️ 100

↪ Data Visualization

a. Bar Chart

-​ A graphical device for DEPICTING categorical data that that has been summarized

-​ Data can be made more obvious with the use of SORTED bar charts
-​ HOWEVER, does not provide exact information of data

b. Pie Chart

-​ A graphical device for PRESENTING summaries based on a subdivision of a circle into sectors,
corresponding to the relative frequency for each class
 -​ Does provide a VISUAL REPRESENTATION
-​ HOWEVER, people do have difficulties piercing differences in areas ( bar chart recommended )

↪ Quantitative Variable Summarization

There are steps necessary to define the classes of a frequency distribution with quantitative data

1.​ Determine the number of NON-OVERLAPPING classes

-​ these classes will SUMMARIZE data into group

-​ should usually be between 5 - 20 classes

> a LARGER number of data ( more than 30 ) should use a LARGER number of classes
> a SMALLER number of data ( less than 30 ) will use 5 -6 classes to summarize data

2.​ Determine the WIDTH of each class

-​ the width of the data should ALWAYS be the same

-​ can be rounded to a WHOLE NUMBER

> calculation = ( largest data value ➖ smallest data value ) ➗ number of classes
3.​ Determine the class LIMIT

-​ Put in place so that each data items BELONGS to only ONE class

> the LOWER CLASS LIMIT is the smallest possible data value

E.g. 10 - 15

> the UPPER CLASS LIMIT is the largest possible data value

E.g. 10 - 15

4.​ Determine the class MIDPOINT

 -​ The value halfway between the LOWER and UPPER class limits

> needs to be put in ascending order

E.g. the class midpoint of the class ‘10 - 15’ is 12.5

This data can be represented with different data visualizations

a. Dot Plot

-​ A graphical device that summarizes data by the number of dots above each data value on the
horizontal access

b. Histogram

-​ A graphical display of a frequency distribution, relative frequency distribution or quantitative data

constructed by;

> placing the class intervals on the horizontal axis

> placing the frequencies, relative frequency or percentage frequencies on the vertical axis

> has NO spaces, unlike a bar chart

> important for providing information about the data DISTRIBUTION

1.​ Can be skewed to the left

-​ The tail extends farther to the left

2.​ Can be skewed to the right

-​ The tail extends farther to the right

3.​ Can be symmetric

-​ The LEFT tail mirrors the shape of the RIGHT tail
↪ Cumulative Distributions

-​ A tabular summary of quantitative data showing the NUMBER of data values that are less than or
equal to the upper class limit of each class

> uses number of classes, class widths and class limits

a.​ Cumulative relative frequency distribution - a cumulative distribution showing the FRACTION
/ PROPORTION of data values

-​ SUM the relative frequencies

-​ DIVIDE the cumulative frequencies by TOTAL ITEMS

b.​ Cumulative percent frequency distribution - a cumulative distribution showing the

PERCENTAGE of data values

-​ MULTIPLY the relative frequencies by 100

↪ Stem and Leaf Display

-​ A graphical display used to show simultaneously the rank order and shape of a distribution of data

> it has 2 primary advantages

1.​ It is EASIER to construct by hand

2.​ Within a class interval, it provides more information than the HISTOGRAM ( shows actual
data )
↪ Summarizing data for TWO variables

a. Crosstabulation

-​ A tabular summary of data for two variables. One class variable is represented by the ROWS, the
other class variable is represented by the COLUMNS

> difficult to decide the number of classes to use when drawing up crosstabulation

➗
1.​ Finds the relative frequency
-​ Frequency total data values

➗ ✖️ 100
2.​ Finds the percentage frequency
-​ ( Frequency total data values )

> a.i. Crosstabulations results in the SIMPSON'S PARADOX

-​ Conclusions drawn from two or more separate crosstabulations that can be reversed when the
data are aggregated into a SINGLE crosstabulation

> two or more SEPARATE crosstabulations can be reversed when aggregated into a single
tabulation

> by UN-AGGREGATING the data - it can show who has a greater record, who has a greater
percentage total etc

-​ When investigating the aggregate or unaggregate forms of the tabulations provide a

BETTER insight + conclusion

( would a hidden variable affect the results - possible providing a different / or better insight
and conclusion ? )

b. Scatter diagram
 -​ A graphical display of the two relationships between two quantitative variables

> uses a trendline - providing an approximation of the relationship between the two variables

> the stronger the relationship , the closer the dots to the trend line

c. Side-by-side Bar Charts

-​ A graphical display for depicting multiple bar charts on the same display

> shows the qualitative data of each class ( good, very good, excellent )

d. Stacked Bar Chart

-​ A bar chart broken into rectangular segments of a different color, showing the relative frequency of
each class
 > shows a clear relationship between the variables

> shows the qualitative data of each class ( good, very good, excellent )

↪ Choosing the Type of Graphical Display

> in order to obtain an accurate DISPLAY and SUMMARY

1.​ For Distribution of Data

Bar chart Histogram

Pie chart Stem and leaf
Dot plot

2.​ For Comparisons

Side by side bar chart Stacked bar chart

3.​ To Show Relationships

Scatter diagram Trendline

> for a COLLECTIVE display, data dashboards are used

( a set of visual displays that ORGANIZE and PRESENT information to monitor the
performance of a company )
 Chapter 3 | Numerical Measures
-​ Involves the 5 number summary

Smallest value, Q1, Median, Q3, Largest value

↪ Measures of Location

a. Mean

-​ A measure of central location computed by summing the data values + dividing by the number of
observations

> denoted by x̄ for samples

> denoted by μ for population

> affected by OUTLIERS ( skews data to left or right )

1)​ Outlier less than mean, skewed to the left ( negative value )
2)​ Outlier more than mean, skewed to the right ( positive value )

3)​ A small amount is SLIGHTLY SKEWED

4)​ A large amount is HEAVILY SKEWED

a i. Weighted Mean

-​ The mean obtained by assigning each observation a WEIGHT that reflects its importance

> ( weight / frequency ✖️ x value ) ➗ total weight / total frequency

a ii. Geometric Mean

-​ A measure of location that is calculated by finding the nth root of the PRODUCT of n values

b. Median

-​ A measure of central location provided by the value in the MIDDLE when the data is arranged in
ascending order
 > if two numbers occur as the median, average those two variables

c. Mode

-​ A measure of location, defines as the value that occurs with the GREATEST frequency

d. Percentiles

-​ A value that provides information about how the data spread over the interval from the SMALLEST
to the LARGEST value

Lp = P / 100 ✖️ ( n + 1 )
> this gives you the POSITION of the percentile
> can be a decimal

P = position 1 ➕ decimal value ✖️ ( position 2 - position 1 )

> the 50th percentile is the MEDIAN

-​ Finds quartiles ( dividing data in four parts )

> includes Q1 , Q2 and Q3

↪ Measures of Variability

-​ How is data SPREAD?

a. Range

-​ A measure of variability, defined to be the largest value MINUS the smallest value

b. Interquartile range

-​ A measure of variability, defined as the difference between the third and the first quartiles

> can be a percentile ( e.g. Q1 is the 25th percentile )

c. Variance

-​ A measure of variability based on the squared deviations of the data values about the mean

> standard deviation is the SQUARE ROOT of these calculations

ci ) Coefficient of variation

-​ A measure of variability computed by dividing the standard deviation by the mean, multiplied by 100
> tells you that the standard deviation is x% of the value of the mean
d. z-Scores

-​ A standardized value, denoting the number of standard deviations is FROM the mean

> shows that the z-score value, x, is x standard deviations BELOW or ABOVE the mean

> negative score is BELOW the mean

> positive score is ABOVE the mean

-​ Can find OUTLIERS

> the first is the lower limit ( LL = Q1 - 1.5 ( IQR ) )

> the second is the upper limit ( UL = Q3 + 1.5 ( IQR ) )

-​ Deals with CHEBYSHEV'S THEOREM

–> At least ( 1 - 1 / z^2 ) of the data values must be within z standard deviations of the mean, where
z is any value greater than 1

a)​ 75% or 0.75 of data MUST be within 2 standard deviations

b)​ 89% or 0.89 of data MUST be within 3 standard deviations
c)​ 94% or 0.94 of data MUST be within 4 standard deviations

↪ the Empirical rule

-​ Used to compute the percentage of data values that must be within 1, 2 and 3
standard deviation of the mean, exhibiting a BELL-SHAPED distribution

> standard deviation = mean + n(standard deviation)

↪ Boxplot

-​ A graphical summary of data based on a five number summary

↪ Measures of Two Variables

a. Covariance

-​ A measure of linear association between two variables. Shows a positive relationship or a negative
relationship

> shows the DIRECTION + LINEAR RELATIONSHIP

> unbounded ( can be any number )

> positive value = positive relationship

> negative value = negative relationship

ai ) Correlation Coefficient

-​ A measure of linear association between two variables of values between -1 and +1

> value NEAR +1 shows a strong positive relationship

> values NEAR -1 shows a strong negative relationship

> values NEAR 0 shows a weak relationship

> an exact value of -1 is a PERFECT negative relationship

> an exact value of +1 is a PERFECT positive relationship

> a BOUNDED value – showing direction + strength

 Chapter 4 | Probability
-​ The likelihood + chance of an event or experiment occurring

↪ Experiments

-​ A process that generates well defined outcomes

E.g. heads, tails, heads, heads, tails

> shows an empirical probability ( the REAL LIFE situations - sample space )
> shows theoretical probability ( ½ chance to get heads on a coin )

-​ Analyzes the LONG TERM pattern

> ‘S’ is the sample point

-​ Important conditions

a)​ 0 =< P(S) =< 1

b)​ EP(S) = 1

↪ Counting techniques

a. Multiple-Step experiments

-​ An experiment that can be described as a sequence

-​ The total number of experimental outcome

> n = n1 x n2 x n3 … x nk

b. Factorial

-​ Arranging ALL experimental outcomes into number of x

>‘❗’
c. Combination

-​ The number of ways n objectives may be selected from among N objects without regard to a
specific order

> from n choose / select k

> nCk

d. Permutations

-​ The number of ways n objectives may be selected from among N objects within a specified order

> from n choose / select k AND arrange k

> the order matters

> nPk
↪ Assigning probability

a. Classical method

-​ A method of assigning probabilities that is appropriate when ALL the experimental outcomes are
equally likely

> equal probability

b. Relative frequency method

-​ Assigning probabilities that is appropriate when data is available, to estimate the proportion of time
the experimental outcome will occur if repeated

> shows probability outcomes in the form of a PERCENTAGE or FRACTION of

c. Subjective method

-​ A guesstimate of an outcome

> not reliable - not used often

↪ Venn diagrams

-​ A graphical representation for showing symbolically the sample space + operations involving
events

a. Union of A and B

-​ The event containing all sample points belonging to A or B or both

> A U B / P(A U B) = P(A) + P(B) - P(A U B)

b. intersection of A and B

-​ The event containing the sample points belonging to both A and B

>A∩B
c. Mutually exclusive events

-​ Events that have no sample points in common

> A ∩ B is empty
> P(A ∩ B) = 0

d. Conditional probability

-​ The probability of an event given that another event already occurred

> probability of A happening, given you know probability of B has occurred

> if independent P(A|B) = P(A) or P(B|A) = P(B)

e. Mutually disjoint

-​ Amounts do NOT meet

> P(A ∩ B) = 0

↪ Events

> a collection of sample points

a. Independent events
-​ Two events that have NO influence on each other

> P(A|B) = P(A)

> P(B|A) = P(B)

b. Dependent events

-​ Where two events are influenced / affected by one another

> P(A|B) ≠ P(A)

> P(B|A) ≠ P(B)
↪ Probability rules

a. Complements of an element

-​ The event consisting of all sample points that are not in the element

> P(A) + P(Ac) = 1

> P(Ac) = 1 - P(A)

b. Addition rule

-​ P(A U B) = P(A) + P(B) - P(A ∩ B)

> P(A ∩ B) is taken twice, so must be removed once

c. Multiplication law

-​ Used to compute the probability of the intersection of two events

> P(A ∩ B) = P(B)P(A | B)

> P(A ∩ B) = P(A)P(B | A)

> For independent events it reduces to P(A ∩ B) = P(A)P(B).

↪ Tree Diagram

-​ A graphical representation that helps in visualizing a multiple-step experiment

 IMPORTANT

A and B are mutually exclusive if and ONLY if P(A n B) = 0

A and B are independent if and ONLY if P(A n B) = P(A) x P(B), or, P(A|B) = P(A), or, P(B|A)
= P(B)

Chapter 5 | Probability Distribution

↪ Random Variables

-​ A numerical description / qualifier of the outcome of an experiment

> QUANTITATIVE in nature

> it is random as it is unknown

> it is a variable as it varies

E.g. amount of people, height, weight

a. Can be a Discrete Random Variable

-​ A random variable that may assume either a finite number of values or an infinite number of values

> can COUNT

b. Can be a Continuous Random Variable

-​ A random variable that may assume any numerical value in an interval / collection of intervals

> when you can MEASURE

IMPORTANT TO NOTE

At least ( >= ) *different for discrete* At most ( =< )

Less than ( < ) Greater than ( > ) *different for discrete*

↪ Probability Distribution

-​ A description of HOW the probabilities are distributed over the values of the random variable

> denoted as f(x)

a. Discrete uniform probability distribution

-​ A probability distribution for which each possible value of the random variable has the SAME
probability

> f(x) = 1/n

> chance is the same
 i) Expected value ii) Variance

-​ The measure of central location -​ A measure of variability of a random

variable
> the MEAN

> finds standard deviation ( the SQUARE

ROOT of this )

↪ Binomial Probability Distribution

a. Binomial experiment

-​ An experiment having four key properties:

1)​ Experiment is a FIXED NUMBER

2)​ Has a SUCCESS or FAILURE
> success = p
> failure = 1 - p
3)​ Trials are INDEPENDENT

-​ Can be used to calculate with a TREE DIAGRAM

> only has two stems as it is a ‘BI’nomial distribution

i) Expected value ii) variance

-​ The average / mean

 Chapter 6 | Continuous Probability
-​ continuous deals with numbers directly after each other ( e.g. 1.000 … 001 )
a. Probability density function
-​ A function used to compute probabilities for a continuous random variable

> it is the area under the graph of the function

b. Uniform probability distribution

-​ A continuous probability distribution for which the probability that the random variable will ASSUME
a value is the same for each interval of equal length

> distributed with a MINIMUM and a MAXIMUM amount

i) Expected value ii) Variance

c. Normal probability distribution

-​ A continuous probability distribution, with its density function as BELL SHAPED and determined
by mean and standard deviation

-​ Has several characteristics

 1)​ The highest point is the MEAN
2)​ The normal distribution is symmetric - reaches INFINITY
3)​ Standard deviation determines how FLAT + WIDE the curve is
4)​ Probabilities are under the curve - E(x) = 1

> has parameters MEAN and VARIANCE

c.i. Standard Normal distribution

-​ A normal distribution with a mean of 0 and standard deviation of 1

i) standard normal random variable

-​ Also known as the z-score ( its standard normal distribution is z ~ N(0,1) )

e. Exponential probability distribution

-​ A continuous probability distribution that is useful in computing probabilities for the time taken to
complete a task

i) cumulative probability

Chapter 7 | Random Variables

-​ Sampling a population brings multiple x-bar + p-bar values that are RANDOM VARIABLES

> in being a random variable it can be distributed

↪ Sampling techniques

a. SRS - simple random sampling

 -​ each element of the population in the sample has an equal chance of being chosen

> uses random function in excel

b. systematic random sampling

-​ Randomly selected sample at a set interval

> e.g. 1 in every 5

c. stratified random sampling

-​ SPLIT population into sub-groups