100% found this document useful (1 vote)
178 views107 pages

MBA 105 Statistical Techniques

MBA 105 Statistical Techniques

Uploaded by

Zalim Zulmi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
178 views107 pages

MBA 105 Statistical Techniques

MBA 105 Statistical Techniques

Uploaded by

Zalim Zulmi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 107

MODULE - 1

Statistical Techniques

NOTES
IMPORTANCE OF STATISTICS IN
BUSINESS MANAGEMENT DOMAIN

CHAPTER OBJECTIVES
1. To familiarize the students with the use of statistics in business domain.
2. To familiarize the students with the data presentation techniques.
3. To develop the quantitative skills of the students to make them skilled at
understanding data, tabulation of data, diagrammatic and graphical
presentation of the data.

CONTENTS
1.1 definition of business statistics.
1.2 importance of statistics in business domain.
1.3 frequency distribution.
1.4 types of frequency distribution.
1.5 diagrammatic presentation of data.
a) Histogram.
b) Frequency polygon.
c) Frequency curve.
d) Cumulative frequency curves (ogive curves).
1.6 graphical presentation of data.
a) Simple bar diagram.
b) Sub divided bar diagram
c) Percentage bar diagram
d) Pie chart

OVERVIEW OF THE CHAPTER


This chapter focuses on the uses of the statistics in the business. It includes
the data organizing methods by using frequency distribution. It also includes the
diagrammatic and graphical representation of data by using various techniques.

KEY WORDS
Business domain, frequency, polygon, Histograms, Ogive curves, pie chart Importance of
Statistics In Business
Management Domain 11
Statistical Techniques INTRODUCTION
Statistics plays an important role in business. Statistical techniques refine
the raw data into finished data such as crude oil is refined into fuel before it can
NOTES be used in the automobile engine. The word Statistics comes from the Italian
word ‘Statista’. It means State. This word was first used by Professor Gottfried
Achenwall in the seventeenth century. As with the many other words the word
Statistics has different meanings for different persons. It is a body of knowledge
or a branch of Mathematics.
The singular word ‘statistic’ means a quantity calculated from the data and
the ‘statistics’ is a subject which includes different theorems and techniques
applicable for analysis of data.

1.1 DEFINITION OF BUSINESS STATISTICS

There have been many definitions of Statistics. A few definitions are given
below.
Merriam Webster’s Definition: Merriam-Webster dictionary defines
statistics as "a branch of mathematics dealing with the collection, analysis,
interpretation, and presentation of masses of numerical data.”
The above definition gives importance to collection of data and presentation
of collected data. It cannot cover all the aspects of Statistics.
Sir Arthur Lyon Bowley’s Definition: Sir Arthur Lyon Bowley defines
statistics as "Numerical statements of facts in any department of inquiry placed
in relation to each other.”
The above definition gives importance for finding the relation between
numerical statements. This definition is very short as it incarcerates the scope of
the Statistics.
Yule and Kendall’s Definition: “By Statistics we mean quantitative data
affected to a marked extent by multiplicity of causes.”
The above definition is also minuscule to explain the huge scope of the
Statistics.
Horace Secrist’s Definition: “By Statistics we mean aggregates of facts
affected to a marked extent by multiplicity of causes, numerically expressed,
enumerated or estimated according to reasonable standards of accuracy, collected
in a systematic manner for a predetermined purpose and placed in relation to
each other.”
The above definition covers intelligibly all the functions and aspects of the
Statistics. It covers almost all the attributes of the Statistics.
The following figure exhibit the meaning of Statistics.
Importance of
Statistics In Business
12 Management Domain
Statistical Techniques

1.2 IMPORTANCE OF STATISTICS IN BUSINESS DOMAIN


NOTES
The Statistical techniques are widely used in Business domain. It provides
a clear picture to business managers for taking the decisions. Diagrammatical
and Graphical presentation clears all the ideas in very short time. Statistical
Methods are used in the business for administration, planning, production,
controlling, marketing research, management, decision-making. The statistical
methods are very useful for the organization for reducing the amount of time,
money, human resource, material and resources.
From last decade of years businesses are growing rapidly all over the world
therefore the competition is also increasing, and the business is becoming the
complex. Hence the entrepreneur’s using the statistics in decision making
process. Now a days Statistics plays a vital role in the decision-making process.
Scope of Statistics in business domain is described with the following heads.
Production: Time series analysis, forecasting techniques are used in the
business for predicting the production capacity. Correlation analysis and
Regression Analysis shows the degree of relation between two variables, such
as price and demand. These two variables are correlated with each other. Price
and demand analysis is useful in production process.
Human resource management: Human resource management is one of
the important task. Human resource manager requires to maintain different
records of the employees as well as organization. Staffing records, payroll,
registers are used in the organization.
Quality Control: Controlling and improving quality has become an
important task in most of the organizations. Quality is a competitive advantage.
A business that can satisfied the customer with continuous improving and
controlling the quality can beat its competitors. Statistical Quality Control deals
with the statistical analysis of the inspection data which is based on sampling
and the principles involved in normal curve.
Financial Accounting: Statistics plays a vital role in creating accounting
reports. These reports are useful in financial decision-making purpose. Statistics
is also useful in accounting to create projections for the next fiscal year.
Market Research: The aim of market research is to identify the gap
between customer needs and products and services being provided in the market.
Statistics provides different research methods for conducting market research
such as qualitative methods, quantitative methods. Qualitative methods deal with
customer profile, background, behaviour and quantitative methods deal with
customers attitude, perception. Survey methods, questionnaire methods are useful
tools for data collection in market research. Importance of
Statistics In Business
Management Domain 13
Statistical Techniques Measures of central tendency and measures of dispersion are the effective
for analysis of data in market research. Diagrammatic and graphical
representations are very effective for comparing the product analysis within short period
NOTES of time.

1.3 FREQUENCY DISTRIBUTION

In Statistics, the frequency (or absolute frequency) of an event “i " is the number “ni”
of times the event occurred in an experiment or study.
A frequency distribution is a table that displays the frequency of various outcomes in
a sample. Each entry in the table contains the frequency or count of the occurrences of
values within a group or interval and in this way the table summarizes the distribution of
values in the sample.
According to Croxton and Cowden “A frequency distribution is a statistical table which
shows the set of all the distinct values of the variable arranged in the order of magnitude
either individually or in groups with their corresponding frequencies side by side.”

1.4 TYPES OF FREQUENCY DISTRIBUTION

(i) Discrete or ungrouped frequency distribution: If the observations are small


and less in numbers, then discrete frequency distribution can be used.
Ex. 1: Given below are the number of batteries sold in 30 working days. Form the
discrete frequency distribution of number of batteries sold.
1, 2, 3, 1, 2, 5, 6, 7, 4, 5, 2, 0, 3, 4, 5, 4, 4, 1, 2, 0, 6, 3, 4, 5, 1, 0, 2, 5, 6, 5.

Solution

Importance of
Statistics In Business
14 Management Domain
Ex. 2: Given below are the marks obtained by 20 students in an Statistical Techniques
examination. Prepare a discrete frequency distribution table.
15, 25, 20, 30, 20, 40, 10, 40, 15, 20, 30, 10,10, 15, 20, 20, 40, 25, 25, 40
NOTES
Solution

(ii) Continuous frequency distribution: If the observations are large in


numbers, then continuous frequency distribution can be used. If the class intervals
are continuous then the frequency distribution is called as continuous frequency
distribution.
Ex. 3: Given below are the marks obtained by 50 students in an
examination. Form a continuous frequency distribution by taking an interval of
10 marks for each class.
16, 10, 12, 14, 20, 18, 40, 42, 35, 28
35, 10, 20, 40, 48, 50, 26, 36, 16, 24
18, 40, 46, 47, 50, 46, 20, 10, 18, 30
55, 60, 68, 70, 56, 58, 70, 62, 50, 18
37, 39, 49, 59, 45, 47, 12, 67, 26, 10

Solution

(ii) Discontinuous frequency distribution: If the class intervals are


discontinuous then the frequency distribution is called as Discontinuous
frequency distribution
Importance of
Statistics In Business
Management Domain 15
Statistical Techniques Ex.4: The following are the salaries of the 45 employees (in thousands of
rupees). Form a discontinuous frequency distribution by taking an interval of 10
for each class.
NOTES 12, 34, 15, 11, 60, 11, 43, 22, 55, 23
33, 11, 21, 45, 68, 23, 16, 16, 26, 24
28, 30, 56, 27, 50, 36, 20, 13, 18, 10
18, 20, 18, 10, 56, 28, 23, 42, 10, 28
31, 19, 29, 59, 45,

Solution

(iv) Cumulative frequency distribution: A distribution containing upper


limits along with less than type cumulative frequency or lower limits along with
more than type cumulative frequency is called as Cumulative frequency
distribution. There are two types of cumulative frequencies Less than cumulative
frequency and More than cumulative frequency.
Less than cumulative frequency: If the frequency of first class interval is
added to the frequency of second class and so on then the frequencies so obtained
are known as Less than cumulative frequency.
More than cumulative frequency: If the total of frequency is taken first and
frequency of first class is subtracted then we get second more than cumulative
frequency. Then if we subtract the frequency of third class from second more
than cumulative frequency then we get third more than cumulative frequency
and so on then the cumulative frequencies so obtained are known as More than
cumulative frequency.
Ex. 5: For the following frequency distribution find (i) less than cumulative
frequency (ii) more than cumulative frequency.

Importance of
Statistics In Business
16 Management Domain
Solution Statistical Techniques

NOTES

Note
(a) It can be noted that, the less than cumulative frequency is increasing
in nature. Less than cumulative frequency of the lowest class is the
same as the usual frequency,
(b) It can be noted that, the more than cumulative frequency is decreasing
in nature. More than cumulative frequency of the lower class is equal
to the total of frequency (i.e. ∑f ) or the less than cumulative frequency
of the highest class.
(v) Relative frequency distribution: A Relative frequency is the
proportion of the number of the observations in the class. The formula of Relative
Frequency is as below.
Class Frequency
Relative Frequency = -------------------------
Total Frequency
Ex. 6: The following data gives the frequency distribution of wages (in
hundreds of rupees) of 60 workers. Find the relative frequencies.

Solution

Importance of
Statistics In Business
Management Domain 17
Statistical Techniques Types of Classification of frequency distribution: There are two types of
classification of frequency distribution inclusive type of frequency distribution
and exclusive type of frequency distribution.
NOTES (a) Inclusive type of frequency distribution: In this type of the
distribution the upper limit of class is not same as the lower limit of
succeeding class. There is discontinuity observed in between the
classes.
e.g.

(b) Exclusive type of frequency distribution: In this type of the


distribution the upper limit of the class is same as the lower limit of
the succeeding class. There is continuity observed in between the
classes.
e.g.

We explain below the three important terms used in frequency distribution.


Class Limits: The class limits are the lowest and the highest values that
can be included in the class. e.g. Take the class (30 – 40). The lowest value of
the class is 30 and the highest value of the class is 40. Therefore, the lower-class
limit is 30 and the upper-class limit is 40.
Class boundaries: The class boundaries are the class numbers up to which
the actual magnitude of observation in the class can broaden e.g. Consider the
classes (30 – 39), (40 – 49). In this case the classes are discontinuous it means
that the observation 39.4 will be rounded off to 39 and placed in the class (30 –
39) whereas the observation 49.1 will be rounded off 49 and placed in the class
(40 – 49). Hence the actual magnitude of the observation in the class (30 – 39)
will be between (29.5 – 39.5)
Class Intervals: The difference between the upper and lower limit of the
class is known as class interval of that class. e.g. Take the class (100 – 200). Here
the class interval is 100 (i.e. 200 minus 100)
Note: Continuous classes are required for graphical representation of data.
When classes are discontinuous then convert it into continuous classes by
subtracting 0.5 from lower class limit and add 0.5 in upper class limit and so on.
Importance of
Statistics In Business
18 Management Domain
e. g. Statistical Techniques

NOTES

1.5 PRESENTATION OF DATA

Mathematical relationship can be represented by graphs and charts. The


relationship between two variables can be shown by graphs and variables are
more then we can use charts.
Graphical Representation of data: Graphs can be represented by the
following methods.
(a). Histogram: Histogram is one of the most popular method used in
practice. It is a continuous graph. There is no any gap between two
bars.
While constructing Histogram the variables are always taken on X- axis
and the corresponding frequencies on the Y- axis. Each class is represented by a
constant continuous interval on X- axis. The frequency of each class is
represented on Y-axis. The frequency represents the height of the bar and the
interval is the breadth of the bar. In this manner, we get the series of the
rectangles.
Ex. 7: Draw a Histogram to represent the following frequency distribution.

Solution

Importance of
Statistics In Business
Management Domain 19
Statistical Techniques Ex. 8: Draw a Histogram to represent the following frequency distribution.

NOTES

Solution

(b) Frequency Polygon: It is another important type of graph. Frequency


polygon enables us to understand the pattern in the data more clearly.
While drawing the frequency polygon mid-points of each class are
taken on X- axis and respective frequencies on Y-axis. After plotting
the points join successive points by line segments. Frequency polygon
is the closed diagram therefore it requires to close at both the ends on
X-axis by taking two more classes. One preceding to first class and
the other succeeding to last class. The figure thus obtained is called as
Frequency Polygon. Frequency polygon can be drawn by two methods.
(i) Using Histogram: Frequency polygon can be drawn by using
Histogram. After drawing Histogram plot the mid points of each class
at the top of each bar. Join successive points by line segments and close
at both the ends on X- axis by taking two more classes. One preceding
to first class and the other succeeding to last class.

Ex. 9: Draw a Frequency Polygon by using Histogram to represent the


following frequency distribution.

Importance of
Statistics In Business
20 Management Domain
Solution Statistical Techniques

NOTES

(ii) Without using Histogram: In this method mid-points of each class


intervals are taken. Take those mid-points on X-axis and respective frequencies
on Y-axis. Plot points by considering mid-points and respective frequencies. Join
successive points by line segments and close at both the ends on X-axis by taking
two more classes. One preceding to first class and the other succeeding to
last class.
Ex. 10: Draw a Frequency Polygon without using Histogram to represent
the following frequency distribution.

Solution

Ex. 11: Draw a Frequency Polygon without using Histogram to represent


the following frequency distribution.

Importance of
Statistics In Business
Management Domain 21
Statistical Techniques Solution

NOTES

(c). Frequency Curve: The procedure of drawing Frequency Curve and


Frequency Polygon without using Histogram is same but there is slightly
difference in it. While drawing frequency curve plotted successive points are
joining by a free hand and in frequency polygon plotted successive points are
joined by line segments (i.e. by using scale).
Ex. 12: Draw a Frequency curve for the following frequency distribution.

Solution

(c) Cumulative Frequency Curves or Ogive Curves: Cumulative


frequency is required for drawing cumulative frequency curves or Ogive Curves.
There are two types of Cumulative Frequency Curves.
(i) Less than Cumulative Frequency Curves: While drawing less than
cumulative frequency curve less than cumulative frequencies are required. Take
upper limits of each class on X-axis and respected less than cumulative
frequencies on Y-axis and plot the points. Join successive points by free hands.
Importance of The figure thus obtained is called as Less than cumulative frequency curve.
Statistics In Business
22 Management Domain
Ex. 13: Draw a Less than cumulative frequency curve for the following Statistical Techniques
frequency distribution.

NOTES

Solution

Solution

(ii) More than Cumulative Frequency Curves: While drawing more


than cumulative frequency curve more than cumulative frequencies are required.
Take lower limits of each class on X-axis and respected more than cumulative
frequencies on Y-axis and plot the points. Join successive points by free hands.
The figure thus obtained is called as More than cumulative frequency curve.
Ex. 14: Draw a More than cumulative frequency curve for the following
frequency distribution.

Importance of
Statistics In Business
Management Domain 23
Statistical Techniques Solution

NOTES

1.6 DIAGRAMMATIC PRESENTATION OF DATA

(a) Simple Bar-Diagram: It is commonly used diagram for presentation


of data. While drawing a Simple bar diagram take discontinuous classes on X-
axis and respective frequencies on Y-axis. Each class interval should be the
breadth of the bar and respective frequency be the height of the bar.
Ex.15: Draw a Simple Bar Diagram for the following data.

Importance of
Statistics In Business
24 Management Domain
Solution Statistical Techniques

NOTES

Ex. 16: Draw a Simple Bar Diagram for the following data.

Solution

(b) Sub – divided Bar Diagram: In Sub divided bar diagram each bar is
divided in same number of parts and each bar representing the magnitude of a
given phenomenon which is sub divided into various components. Each
component of a bar occupies a part of the bar proportional to its share in the total.
Ex. 18: Draw a Sub Divided Bar Diagram for the following data.

Solution

Importance of
Statistics In Business
Management Domain 25
Statistical Techniques Ex. 17: Draw a Sub Divided Bar Diagram for the following data.

NOTES

Solution

(c) Percentage Bar Diagram: In percentage Bar Diagram length of each


bar is kept equal to 100 and segments are divided in these bars to represent the
components of the aggregate.
Ex. 18: Represent the following by Percentage Bar Diagram for the
following data.

Solution

Importance of
Statistics In Business
26 Management Domain
Ex. 19: Represent the following by Percentage Bar Diagram for the Statistical Techniques
following data.

NOTES

Solution

(d) Pie Chart: Pie diagrams are commonly used in practice. Percentages
are used for showing the breakdowns. While constructing the pie-chart follow
the following procedure.

Calculate percentage of given data.


Convert the percentages into angles by using the formula:
Percentage Outlay --------------------- X 360
100
Ex.20. Represent the following data by Pie Chart.

Solution

Importance of
Statistics In Business
Management Domain 27
Statistical Techniques SUMMARY
In this chapter, we have studied the definitions of Statistics, importance of
Statistics in Business, Organising the data using frequency distribution and types
NOTES of frequency distribution. The chapter also elaborates various graphical
representation methods such as Histogram, frequency polygon and frequency
curves, cumulative frequency curves and diagrammatic representation methods
as Simple bar diagram, sub-divided bar diagram, percentage bar diagram and
pie-chart.

REFERENCES
1. "Definition of Statistics". www.merriam-webster.com. Retrieved 2016-
05-28. Web on 3 Nov 2017 4.30 pm in Wikipedia.
2. "Essay on Statistics: Meaning and Definition of Statistics". Economics
Discussion. 2014-12-02. Retrieved 2016-05-28. Web on 3 Nov 2017
4.30 pm in Wikipedia.
3. S.P. Gupta, “Statistical Methods.” Sultan Chand and Sons, Thirteenth
Revised Edition.2001.
4. M.G. Dhaygude,” Statistical and Quantitative Methods” Everest
Publishing House2004.Print.

EXERCISE
1. Define Statistics and explain its importance in Business.
2. “Statistics play an important role not only in production but also in
human resource management”. Explain this statement with appropriate
examples.
3. “Statistics is an effective tool in marketing”. Comment on the
statement.
4. “Successive marketing managers are those who are well equipped with
statistical tools.” Critically examine the statement and justify your
stand in favour as well as against the statement.
5. What is Statistics? Explain how Statistics is useful in decision making?
6. “Graphs and diagrams are effective methods than other methods of
presenting the data.” Justify your answer.
7. What is Histogram? How do you construct it? Explain with suitable
example.
8. “Pie-diagrams are useful for representing the percentages. Illustrate
your answer with suitable example.
9. Write note on Cumulative frequency distribution.
10. Draw a Histogram and frequency polygon for the following data.

Importance of
Statistics In Business
28 Management Domain
11. Construct a frequency table for the following data regarding annual Statistical Techniques
profits in thousands of rupees of 40 firms. Take classes as 10 – 20,
20 – 30…
11, 37, 25, 21, 60, 11, 43, 22, 25, 53 NOTES
33, 11, 31, 65, 68, 73, 16, 16, 86, 74
28, 20, 56, 27, 50, 39, 20, 13, 18, 70
18, 70, 18, 10, 56, 28, 13, 32, 10, 88

12. Given below are the weekly wages in rupees of 50 workers in a factory.
Prepare a frequency distribution table with classes as 10 – 19, 20 – 29, .....
58, 65, 47, 71, 67, 41, 55, 28, 27, 59
34, 16, 37, 68, 68, 73, 36, 56, 59, 33
28, 20, 56, 17, 50, 39, 20, 13, 18, 70
18, 30, 18, 10, 56, 28, 13, 36, 60, 58
61, 66, 69, 81, 60, 12, 37, 12, 45, 48

13. Draw a Histogram and frequency polygon for the following data.

14. Draw a frequency polygon for the following data.

15. Draw cumulative frequency curves for the following data.

16. The following data is related to the expenditure of a family. Represent


the data by using Pie- diagram.

17. Represent the following data by an appropriate diagram.

Importance of
Statistics In Business
***** Management Domain 29
MODULE - 2
Statistical Techniques

NOTES
MEASURES OF CENTRAL TENDENCY
AND DISPERSION

CHAPTER OBJECTIVES
1. To determine the central tendency and dispersion for Individual and
Ungrouped data.
2. To determine the central tendency and dispersion for Grouped data.

CONTENTS
A measures of central tendency
2.1: arithmetic mean,
2.2: median,
2.3: mode
2.4: partition values: quartiles, deciles and percentiles

B measures of dispersion:
2.5: range,
2.6: mean deviation,
2.7: variance and standard deviation,
2.8: coefficient of variation

KEY WORDS
Central tendency, Dispersion, Partition values

OVERVIEW OF THE CHAPTER


This chapter focuses on the measures of central tendency such as mean,
median and mode. It also includes the measures of dispersion such as Range,
Mean Deviation, Standard Deviation, Coefficient of Variation.

A. MEASURES OF CENTRAL TENDENCY


In previous chapter, we studied the representation of data by graphical
methods. In analysis purpose, most of the times one representative value is
required instead of whole data or observations. In other words, the collected data
is very wide and hence it is necessary to find a representative figure for the data.
There is a tendency of the collected data to concentrate about a value, called as
Central Tendency and that value is called as measure of central tendency.
Measures of The mostly used measures of central tendency are: Mean, Median and
Central Tendency Mode.
30 And dispersion
Statistical Techniques

2.1 ARITHMETIC MEAN


NOTES
The arithmetic mean, μ or simply called mean, is obtained by adding
together all the measurements and dividing by the total number of measurements
taken. Mathematically it is given as

Where - for grouped data: fi - is the frequency in the ith class,


xi - is the class mark in the ith class;
for ungrouped data: fi - is the frequency in the ith datum,
xi - is the value in the ith datum.
Arithmetic mean can be used to calculate any numerical data and it is always
unique. It is obvious that extreme values affect the mean. Also, arithmetic mean
ignores the degree of importance in different categories of data.
Example 2: Given the following set of ungrouped data:
20, 18, 15, 15, 14, 12, 11, 9, 7, 6, 4, 1
Find the mean of the ungrouped data.

Weighted Arithmetic Mean


In order to consider the importance of some data, different weighting
factors, wi , can be assigned to individual datum. Hence the weighted arithmetic
mean, μ, is given as:

Where wi is the weight for the ith datum.


fi and xi are defined same as those in the arithmetic mean for
ungrouped and grouped data.
Measures of
Central Tendency
And dispersion 31
Statistical Techniques

2.2 MEDIAN
NOTES
Median is defined as the middle item of all given observations arranged in
order. For ungrouped data, the median is obvious. In case of the number of
measurements is even, the median is obtained by taking the average of the
middle.
Example 3: The median of the ungrouped data::
20, 18, 15, 15, 14, 12, 11, 9, 7, 6, 4, 1 is

Median =

Median = 11.5
For grouped data, the median can be found by first identify the class
containing the median, then apply the following formula:

where: l1 is the lower class boundary of the median class;


n is the total frequency;
C is the cumulative frequency just before the median class;
fm is the frequency of the median;
l2 is the upper class boundary containing the median.
It is obvious that the median is affected by the total number of data but is
independent of extreme values. However if the data is ungrouped and numerous,
finding the median is tedious. Note that median may be applied in qualitative
data if they can be ranked.

2.3 MODE

Mode is the value which occurs most frequency. The mode may not exist,
and even if it does, it may not be unique.
For ungrouped data, we simply count the largest frequency of the given
value. If all are of the same frequency, no mode exits. If more than one values
have the same largest frequency, then the mode is not unique.
Example 4: The value for the mode of the data in Example
Measures of 3 is 15 (unimodal)
Central Tendency
32 And dispersion
Example 5 : Calculate mode for the following data. Statistical Techniques
{2, 2, 2, 4, 5, 6, 7, 7, 7}
Mode = 2 or 7 (Bimodal)
NOTES
For grouped data, the mode can be found by first identify the largest
frequency of that class, called modal class, then apply the following formula on
the modal class:

where: l1 is the lower class boundary of the modal class;


fa is the difference of the frequencies of the modal class
with the previous class and is always positive;
fb is the difference of the frequencies of the modal class
with the following class and is always positive;
l2 is the upper class boundary of the modal class.
Geometrically the mode can be represented by the following graph and can
be obtained by using similar triangle properties. The formula can be derived by
interpolation using second degree polynomial.

Note that the mode is independent of extreme values and it may be applied
in qualitative data.
For symmetrically distributed data, the mean, median and mode can be used
almost interchangeably.

. (Mean - Median)
For moderately skewed distribution data, their relationship can be given by
Mean - Mode ≈ 3
Physically, mean can be interpreted as the center of gravity of the
distribution. Median divides the area of the distribution into two equal parts and Measures of
mode is the highest point of the distribution. Central Tendency
And dispersion 33
Statistical Techniques

2.4 DECILES, PERCENTILE, AND QUARTILES


NOTES
Decile divides the distribution into ten equal parts while percentile divides
the distribution into one hundred equal parts. There are nine deciles such that
10% of the data are D1; 20% of the data are D2; and so on. There are 99
percentiles such that 1% of the data are P1; 2% of the data are P2; and so
on. Fractile, even more flexible, divides the distribution into a convenience
number of parts. Quartiles are the most commonly used values of position
which divides distribution into four equal parts such that 25% of the data are
Q1; 50% of the data are Q2; 75% of the data are Q3. The first quarter is
conventionally denoted as Q1, while the second and third quarters grouped
together is Q2 and the last quarter is Q3. Note that Q2 includes the median,
contains half of the frequency and excludes extreme values. It is also de

B. MEASURES OF DISPERSION
Here are some definitions of Dispersion as given by Statisticision from time
to time.
“Dispersion is the measure of variation of the item”- A.L. Bowley.
“The degree to which the numerical data tend to spread about an average
value is called the variation or dispersion of the data.” – Spiegal.
The measures of Dispersion are range, mean absolute deviation, variance,
standard deviation, coefficient of variation

2.5 RANGE

Range is the difference between two extreme values. The range is easy to
calculate but cannot be obtained if open ended grouped data are given. noted the
value (Q3 - Q1) / 2 as the Quartile Deviation, QD, or the semi-interquartile range.

2.6 MEAN ABSOLUTE DEVIATION

Mean absolute deviation is the mean of the absolute values of all deviations
from the mean. Therefore, it takes every item into account. Mathematically it
is given as:

where: fi is the frequency of the ith item;


Measures of xi is the value of the ith item or class mark;
Central Tendency
34 And dispersion μ is the arithmetic mean.
Statistical Techniques

2.7 VARIANCE AND STANDARD DEVIATION


NOTES
The variance and standard deviation are two very popular measures of
variation. Their formulations are categorized into whether to evaluate from a
population or from a sample.
The population variance, μ2, is the mean of the square of all deviations from
the mean. Mathematically it is given as:

where: fi is the frequency of the ith item;


xi is the value of the ith item or class mark;
μ is the population arithmetic mean.
The population standard deviation σ is defined as σ =
The sample variance, denoted as s2 gives:

where: fi is the frequency of the ith item;


xi is the value of the ith item or class mark;
x is the sample arithmetic mean.
The sample standard deviation, s, is defined as s = .
Note that when calculating the sample variance, we have to subtract 1 from
the total frequency which appears in the denominator. Although when the total
frequency is large, s ≈ σ, the subtraction of 1 is very important.

Example 6: For the following data calculate


(i) Mean and Median, Q1 and Q3
(ii) Mode and Standard deviation
Gas Consumption Frequency
10 – 19 1
20 – 29 0
30 – 39 1
40 – 49 4
50 – 59 7
60 – 69 16
70 – 79 19
80 – 89 20
90 – 99 17
100 – 109 11
110 – 119 3
120 – 129 1 Measures of
Central Tendency
100 And dispersion 35
Statistical Techniques

NOTES

2.8 COEFFICIENT OF VARIATION

The coefficient of variation is a measure of relative importance. It does not


depend on unit and can be used to make comparison even two samples differ in
means or relate to different types of measurements.
The coefficient of variation gives

Example 7

Measures of
Central Tendency
36 And dispersion
Statistical Techniques

NOTES

Skewness
The skewness is an abstract quantity which shows how data piled-up. A
number of measures have been suggested to determine the skewness of a given
distribution. One of the simplest one is known as Pearson’s measure of skewness:

If the tail is on the right, we say that it is skewed to the right, and the
coefficient of skewness is positive.

If the tail is on the left, we say that is skewed to the left and the coefficient
of skewness is negative.

Measures of
Central Tendency
And dispersion 37
Statistical Techniques Example 12: The salary scales of the two companies are given below.
Company A: $5,000 $15,000 $25,000 $35,000 $45,000 $55,000
Company B: $5,000 $5,000 $5,000 $55,000 $55,000 $55,000
NOTES
Calculate: (i) Range, Mean Absolute Deviation, Variance
(ii) Standard deviation, Coefficient of Variation, Skewness
(i) Range
Company A: $55,000 $5,000
= $50,000
Company B: $55,000 $5,000
= $50,000

(ii) Mean absolute deviation


Company A: $ ( |5,000 30,000| + |15,000 30,000| + |25,000
30,000| + |35,000 30,000| + |45,000 30,000| +
|55,000 30,000| ) / 6
= $15,000

Company B: $ ( |5,000 30,000| + |5,000 30,000| + |5,000


30,000| + |55,000 30,000| + |55,000 30,000| +
|55,000 30,000| ) / 6
= $25,000

(iii) Variance
Company A: $2 { (5,000 30,000)2 + (15,000 30,000)2 + (25,000
30,000)2 + (35,000 30,000)2 + (45,000 30,000)2 +
(55,000 30,000)2 } / 6
= $2291,666,667
Company B: $2 { (5,000 30,000)2 + (5,000 30,000)2 + (5,000
30,000)2 + (55,000 30,000)2 + (55,000 30,000)2 +
(55,000 30,000)2 } / 6
= $2625,000,000

(iv) Standard deviation


Company A: $√ 291,666,667
= $17,078
Company B: $√ 625,000,000
Measures of
Central Tendency = $25,000
38 And dispersion
(v) Coefficient of variation Statistical Techniques
Company A: $17,078 / $30,000 100%
= 56.93%
NOTES
Company B: $25,000 / $30,000 100%
= 83.33%

SUMMARY
In this chapter, we have studied the measures of central tendency as mean,
median and mode. The examples of individual data, ungrouped data and grouped
data are used for calculation of various central tendencies. Measures of variation
such as range, mean deviation, standard deviation, and coefficient of variation
are also studied.

REFERENCES
1. www.mypolyuweb.hk/machanck/lectnotes/c1_desc.doc

EXERCISE
1. In the following list, post a D for the situations in which statistical
techniques are used for the purpose of description and an I for those
in which the techniques are used for the purpose of inference.
(a) The price movements of 50 issues of stock are analysed to
determine whether stocks in general have gone up or down during
a certain period.
(b) A statistical table is constructed for presenting the passenger-
miles flown by various commercial airlines in the United States.
(c) The average of a group of test scores is computed so that each
score in the group can be classified as being either above or below
average.
(d) Several manufacturing firms in a industry are surveyed for the
purpose of estimating industrywide investment in capital
equipment.

2. No matter how few elements are included in a statistical population,


however, a sample taken from that population (can/cannot) be larger
than the population itself.

3. Thus any descriptive measurement of a population is considered to be


a (statistics/parameter), and a descriptive measurement of a sample is
a sample ________.

Measures of
Central Tendency
And dispersion 39
Statistical Techniques 4. The word “statistics” has at least three distinct meanings, depending on
the context in which it is used. It may refer to:
(i) the procedure of statistical analysis
NOTES
(ii) descriptive measures of a sample
(iii) the individual measurements, or elements, that make up either a
sample or a population.
(a) When one becomes “an accident statistics” by being included in
some count of accident frequency, the term is used in the sense
of definition ____.
(b) According to the definitions in a course of study called “Business
Statistics” the term “statistics” is usually used in the sense of
definition ____.
(c) According to the definitions when such sample statistics as the
proportion of a sample in favour of a proposal and the average
age of those in the sample are determined, the term “statistics” is
being used in the sense of definition ____.

5. The two major applications of the tools of statistical analysis are


directed toward the purposes of statistical ____________ and
statistical ____________.

6. When all the elements in a statistical population are measured, the


process is referred to as “taking a __________“. If only a portion of
the elements included in a statistical population is measured, the
process is called ____________.

7. Which of the following measures of variability is not dependent on


the exact value of each observation?
(a) range
(b) variance
(c) standard deviation
(d) coefficient of variation

8. A measure of dispersion which is insensitive to extreme values in the


data set is the:
(a) Quartile deviation
(b) Standard deviation
(c) Average absolute deviation
(d) All of the above
Measures of
Central Tendency
40 And dispersion
9. An absolute measure of dispersion which expresses variation in the same Statistical Techniques
units as the original data is the:
(a) Standard deviation
NOTES
(b) Coefficient of variation
(c) Variance
(d) All of the above

10. How does the computation of a sample variance differ from the
computation of a population variance?
(a) μ is replaced by x
(b) N is replaced by n 1
(c) N is replaced by n
(d) a and c but not b
(e) a and b but not c

11. Which measure of central tendency would be most useful in each of


the following instances?
(a) The production manager for a manufacturer of glass jars, who is
concerned about the proper jar size to manufacture, has sample
data on jar sizes ordered by customers. Would the mean, median,
or modal jar size be of most value to the manager?
(b) The sales manager for a quality furniture manufacturer is
interested in selecting the regions most likely to purchase his
firm’s products. Would he be most interested in the mean or
median family income in prospective sales areas?
(c) A security analyst is interested in describing the daily market
price change of the common stock of a manufacturing company.
Only rarely does the market price of the stock change by more
than one point, but occasionally the price will change by as many
as four points in one day. Should the security analyst describe
the daily price change of the stock in terms of the mean, median,
or modal daily market price change?

12. Why isn’t an average computed from a group frequency distribution


exactly the same as that computed from the original raw data used to
construct the distribution?

13. For which type of distribution (positively skewed, negatively skewed,


or symmetric) is:
(a) The mean less than the median?
(b) The mode less than the mean?
(c) The median less than the mode? Measures of
Central Tendency
And dispersion 41
Statistical Techniques 14. The following scores represent the final examination grade for an
elementary statistics course:
23 60 79 32 57 74 52 70 82 36
NOTES 80 77 81 95 41 65 92 85 55 76
52 10 64 75 78 25 80 98 81 67
41 71 83 54 64 72 88 62 74 43
60 78 89 76 84 48 84 90 15 79
34 67 17 82 69 74 63 80 85 61
Using 10 class intervals with the lowest starting at 9:
(a) Set up a frequency distribution.
(b) Construct a cumulative frequency distribution.
(c) Construct a frequency histogram.
(d) Construct a smoothed cumulative frequency polygon.
(e) Estimate the number of people who made a score of at least 60
but less than 75.
(f) Discuss the skewnessF of the distribution.

15. Classify the following random variables as discrete or continuous.


(a) The number of automobile accidents each year in Hong Kong.
(b) The length of time to do problem 1 above.
(c) The amount of milk produced yearly by a particular cow.
(d) The number of eggs laid each month by 1 hen.
(f) Numbers of shares sold each year in the stock market.
(d) The weight of grain in kg produced per acre.

16. An electronically controlled automatic bulk food filler is set to fill tubs
with 60 units of cheese. A random sample of five tubs from a large
production lot shows filled weights of 60.00, 59.95, 60.05, 60.02 and
60.01 units. Find the mean and the standard deviation of these fills.

17. In four attempts, it took a person 48, 55, 51 and 50 minutes to do a


certain job.
(a) Find the mean, the range, and the standard deviation of these four
sample values.
(b) Subtract 50 minutes from each of the times, recalculate the mean,
the range, and the standard deviation, and compare the results
with those obtained in part (a).
(c) Add 10 minutes to each of the times, recalculate the mean, the
Measures of
Central Tendency range, and the standard deviation, and compare the results with
42 And dispersion those obtained in part (a).
(d) Multiply each of the sample values by 2, recalculate the mean, Statistical Techniques
the range, and the standard deviation, and compare the results
with those obtained in part (a).
(e) In general, what effect does (1) adding a constant to each sample NOTES
value, and (2) multiplying each sample value by a positive
constant, have on the mean, the range, and the standard deviation
of a sample?

18. Find the mean, median and mode for the set of numbers
(a) 3, 5, 2, 6, 5, 9, 5, 2, 8, 6;
(b) 51.6, 48.7, 50.3, 49.5, 48.9.

19. The lengths of a large shipment of chromium strips have a mean of


0.44 m and standard deviation of 0.001 m. At least what percentage
of these lengths must lie between
(a) 0.438 and 0.442 m?
(b) 0.436 and 0.444 m?
(c) 0.430 and 0.450 m?

20. The 1971 populations and growth rates for various regions are given
below. Find the growth rate for the world as a whole
Region Population (millions) Annual Growth Rate (%)
Europe 470 0.8
USSR 240 1.1
N. America 230 1.3
Oceania 20 2.1
Asia 2,100 2.3
Africa 350 2.6
S. America 290 2.9

21. Suppose that the annual incomes of the residents of a certain country
has a mean of $48,000 and a median of $34,000. What is the shape of
the distribution?

22. In a factory, the time during working hours in which a machine is not
operating as a result of breakage or failure is called the ‘downtime”.
The following distribution shows a sample of 100 downtimes of a
certain machine (rounded to the nearest minute) :
Downtime Frequencies
0-9 3
Measures of
10 - 19 13 Central Tendency
And dispersion 43
Statistical Techniques 20 - 29 30
30 - 39 25
40 - 49 14
NOTES
50 - 59 8
60 - 69 4
70 - 79 2
80 - 89 1
With reference to the above distribution, calculate
(a) the mean.
(b) the standard deviation.
(c) the median.
(d) the quartiles Q1 and Q3.
(e) the deciles D1 and D9.
(f) the percentiles P5 and P95.
(g) Pearson’s first and second coefficients of skewness.
(h) the modal downtime of the distribution by the empirical formula
(using the results obtained in part (a) and part (c) only). Compare
this result with the mode obtained in part (g).

23. Consider the following frequency distribution of weights of 150 bolts


Weight (grams) Frequency
5.00 and less than 5.01 4
5.01 and less than 5.02 18
5.02 and less than 5.03 25
5.03 and less than 5.04 36
5.04 and less than 5.05 30
5.05 and less than 5.06 22
5.06 and less than 5.07 11
5.07 and less than 5.08 3
5.08 and less than 5.09 1
(a) Calculate the mean and standard deviation of the weights of bolts
to three decimal places.
(b) Estimate the number of bolts which are within one standard
deviation of the mean.
(c) Suppose that each bolt has a nut attached to it to make a nut-and-
bolt. Nuts have a distribution of weights with a mean of 2.043
Measures of
Central Tendency grams and standard deviation 0.008. Calculate the standard
44 And dispersion deviation of the weights of nut-and-bolts.
24. A random sample of 11 vouchers is taken from a corporate expense Statistical Techniques
account. The Voucher amounts are as follows:
$276.72 194.17 259.83 249.45
201.43 237.66 199.28 211.49 NOTES
240.16 261.10 226.21
Compute
(a) the range;
(b) the interquartile range;
(c) the variance (definitional and computational);
(d) the standard deviation;
(e) the coefficient of variation.

25. A hardware distributor reports the following distribution of sales from


a sample of 100 sales receipts.
Dollar Values Number of
of Sales Sales (f)
$ 0 but less than 20 16
20 but less than 40 18
40 but less than 60 14
60 but less than 80 24
80 but less than 100 20
100 but less than 120 8
Total 100
Find :
(a) the variance (definitional and computational).
(b) the standard deviation
(c) the coefficient of variation

26. The National Space Agency requires that all resistors used in electronic
packages assembled for space flight have a coefficient of variation less
than 5 percent. The following resistors made by the Mary Drake
Company have been tested with results as follows :
Resistor Mean Resistance Standard Deviation
A 100 K-ohms 4 K-ohms
B 200 12
C 300 14
D 400 16
E 500 18
F 600 20
Which of the resistors meets specifications?
27. Salaries paid last year to supervisors had a mean of $25,000 with a
standard deviation of $2000. What will be the new mean and standard
deviation if all salaries are increased by $2500? Measures of
Central Tendency
***** And dispersion 45
MODULE - 3
Statistical Techniques

NOTES
CORRELATION ANALYSIS

CHAPTER OBJECTIVE
1. To familiarize the students with one of the technique of analysis of two
variables.
2. To develop the quantitative skills of the students to make them skilled at
understanding the bivariate data analysis technique.

CONTENTS
3.1 definition of correlation.
3.2 types of correlation.
3.3 Methods of studying correlation.
A) scatter diagrams.
B) karl pearson’s coefficient of correlation.
C) rank correlation.
3.4 Coefficient of determination.

OVERVIEW OF THE CHAPTER


This chapter focuses on the one of the method of analysis of bivariate data.
It includes the meaning of Correlation and various methods of studying
Correlation such as Scatter diagrams, Karl Pearson’s Coefficient of Correlation.

KEY WORDS
Correlation, Scatter Diagram, Rank Correlation, Determination.

INTRODUCTION
If for two quantities under study variation in one is accompanied by the
variation in the other, then we say that these quantities are correlated. The degree
of relationship between the variables is measured through the correlation analysis.
Therefore, the Correlation is an analysis of the covariation between two or more
variables. The measure of correlation is called as correlation coefficient. It
indicates the direction and degree of correlation between the variables.

46 Correlation Analysis
Statistical Techniques

3.1 DEFINITION OF CORRELATION


NOTES
Correlation is any of a broad class of statistical relationships involving
dependence, though in common usage it most often refers to how close two
variables are to having a linear relationship with each other. Familiar examples
of Correlation includes the correlation between the physical statures of parents
and their offspring, and the correlation between the demand for a product and
its price.
According to the Oxford Dictionary the Correlation means a mutual
relationship or connection between two or more things.

3.2 TYPES OF CORRELATION

There are basically three types of Correlation


1. Positive Correlation
2. Negative Correlation
3. No Correlation

1. Positive Correlation
If both the variables under study vary in the same direction i.e. if one
variable (say X) increases then the other variable (say Y) also increases
correspondingly or vice versa then there is a Positive Correlation between the
two variables.
e.g. Rainfall and yield of a crop, consumer spending and gross domestic
product (GDP). Demand and price etc.

2. Negative Correlation
If variables vary in opposite direction i.e. if one variable (say X) increases
then the other variable (say Y) decreases correspondingly or vice versa then there
is a Negative Correlation between the two variables.
e.g. Price and demand, the more time a person spends at the mall purchasing
goods, the less money he has in his . The higher an investor's mutual fund , the
lower his investment returns. The more hours a person spends at the office, the
less time he has for other activities.

3. No Correlation
If change in one variable is not related to change in other variable then there
is no Correlation between two variables.

Correlation Analysis 47
Statistical Techniques There are other types of Correlation as:
• Simple Correlation
• Partial Correlation
NOTES • Multiple Correlation
• Linear Correlation
• Non- Linear Correlation
Simple Correlation: If two variables are study simultaneously then there
is Simple Correlation between two variables.
Partial Correlation: If more than two variables are study and considering
only two variables at a time and treating remaining variables constants then there
is Partial Correlation between variables.
Multiple Correlation: If more than two variables are study simultaneously
then there is Multiple Correlation between variables.
Linear Correlation: If a unit change in one variable then there is a constant
corresponding change in other over the entire range of values then there is Linear
Correlation between the variables.

3.3 METHODS OF STUDYING CORRELATION

The methods of studying correlation are as follows.


A. Scatter Diagram
B. Karl Pearson’s Coefficient of Correlation
C. Rank Correlation Coefficient

A. Scatter Diagram
It is a graphical representation of bivariate distribution.
Let there be n pairs of values representing the two variables X and Y. i.e.
(x1, y1), (x2, y2)……., (xn, yn),. Taking one of the variable on X axis and other
on Y axis we get a diagram called as Scatter Diagram. This diagram explains the
nature of Correlation. But we can’t establish the exact degree of Correlation here.
These diagrams are classified into five different categories which are
shown below.

LINEAR CORRELATION
The purpose of a linear correlation analysis is to determine whether there is
a relationship between two sets of variables. We may find that: 1) there is a
positive correlation, 2) there is a negative correlation, or 3) there is no correlation.
These relationships can be easily visualized by using scatter diagrams.

48 Correlation Analysis
Positive Correlation Statistical Techniques

NOTES

Notice that in this example as the heights increase, the diameters of the
trunks also tend to increase. If this were a perfect positive correlation all of the
points would fall on a straight line. The more linear the data points, the closer
the relationship between the two variables.

Negative Correlation

Notice that in this example as the number of parasites increases, the harvest
of unblemished apples decreases. If this were a perfect negative correlation all
of the points would fall on a line with a negative slope. The more linear the data
points, the more negatively correlated are the two variables.

Correlation Analysis 49
Statistical Techniques No Correlation

NOTES

HOW TO DECIDE IF THERE IS A LINEAR CORRELATION OR NOT


First, locate the samples size nearest to yours in the table below at look at
the critical region for that samples size:
If your correlation coefficient (r-value) falls within the critical region, your
two variables do not change in a regular enough way to be considered significant.
You conclude that your data indicate no correlation.
If your correlation coefficient is a larger positive number than the critical
region, your data indicate a significant positive correlation. The closer your r-
value is to 1.0, the stronger the correlation.
If your correlation coefficient is a larger negative number than the critical
region, your data indicate a significant negative correlation. The closer your r-
value is to —1.0, the stronger the negative correlation.

B. Karl Pearson’s Coefficient of Correlation


Karl Pearson’s Coefficient of Correlation is denoted by r and it is defined
as follows:

r=

Ex 1. Calculate Coefficient of Correlation for the following data.

50 Correlation Analysis
Solution Statistical Techniques

NOTES

Here, n = 9, X X = ∑ x / n = 154 / 9 = 17.11


Y = ∑ y / n = 247 / 9 = 27.44

r=

r = 4382 – 9 (17.11) (27.44) /√ (2816 – 9(14.7)2. (7317 – 9(27.44)2


r = 0.2281

Correlation by Assuming mean method


Ex 2:. Calculate Coefficient of Correlation for the following data.

Solution

Correlation Analysis 51
Statistical Techniques

NOTES

52 Correlation Analysis
Ex 3: Calculate Coefficient of Correlation for the following data. Statistical Techniques

NOTES

Solution

By using Karl Pearson’s assumed mean method


r = - 0.2

b) Rank Correlation Coefficient


Spearman’s Rank Correlation Coefficient
The Rank Correlation test is a method of comparing two sets of data.The
data you will be comparing are called variables.
In the equation:
R= 1- 6Σd 2
R = The rank correlation coefficient
n3 – n Σ = the sum of
n = the number of variables being correlated

A) When ranks are given


Ex: 4: Eight brands of lipsticks are ranked by two Judges in a competition.
The best brand of lipsticks has rank 1.

Correlation Analysis 53
Statistical Techniques Calculate Spearman’s rank coefficient for above data and comment on
result.

NOTES Solution
Let us consider the ranks given by Judge I and Judge II be X and Y
respectively.

Here R = -0.5 Therefore there is exactly opposite approach of judging


between two judges.

B) When ranks are not given


When direct observations are given then rank each observation by rank. The
least observation will have rank 1. The second least observation will have rank
2 and so on. Sometimes more than one items are given the same rank. In this
case new ranks are allocated to the items that have tied ranks.

54 Correlation Analysis
Ex. 5: The marks of 12 candidates in a certain test of Statistics and Statistical Techniques
Economics are as follows;

NOTES

Calculate Spearman’s rank correlation coefficient.

Solution
First, we must rank the data.
Geography History
19 = 1 13 = ½ (1+2) = 1.5
18 = 1/3 (2 + 3 + 4) = 3 12 = 1/3 (3+4+5) = 4
17 = ½ (5 + 6) = 5.5 11= 1/3 (6+7+8) = 7
16 = ½ (7 + 8) = 7.5 10 = 9
15 = ½ (9 + 10) = 9.5 9 = 10
14 = 11 8 = 11
10 = 12 7 = 12

Let the ranks given to Statistics be X and the ranks given to Economics be Y

Correlation Analysis 55
Statistical Techniques

NOTES

Hence, there is positive correlation between the Statistics and


Economics results.

3.4 COEFFICIENT OF DETERMINATION

In statistics, the coefficient of determination, denoted R2 or r2 and


pronounced "R squared", is the proportion of the variance in the dependent
variable that is predictable from the independent variable(s).
It is a statistic used in the context of statistical models whose main purpose
is either the prediction of future outcomes or the testing of hypotheses, on the
basis of other related information. It provides a measure of how well observed
outcomes are replicated by the model, based on the proportion of total variation
of outcomes explained by the model.
There are several definitions of R2 that are only sometimes equivalent. One
class of such cases includes that of simple linear regression where r2 is used
instead of R2. When an intercept is included, then r2 is simply the square of the
sample correlation coefficient (i.e., r) between the observed outcomes and the
observed predictor values. If additional are included, R2 is the square of the
coefficient of multiple correlation. In both such cases, the coefficient of
determination ranges from 0 to 1.
Important cases where the computational definition of R2 can yield negative
values, depending on the definition used, arise where the predictions that are
being compared to the corresponding outcomes have not been derived from a
model-fitting procedure using those data, and where linear regression is
conducted without including an intercept. Additionally, negative values of R2
may occur when fitting non-linear functions to data. In cases where negative
values arise, the mean of the data provides a better fit to the outcomes than do
the fitted function values, according to this criterion.

56 Correlation Analysis
Coefficient of Determination = r2 = Explained Variance / Total Variance Statistical Techniques
Ex: 6: Calculate coefficient of correlation, coefficient of determination from
the following data.
NOTES

Solution

Here r = 0.99
R2 = 0.9801

SUMMARY
In this chapter, we have studied the definitions of Correlation, types of
correlation. This chapter also elaborates methods of studying correlation such as
scatter diagrams, Karl Pearson’s coefficient of correlation, Spearman’s rank
correlation. Coefficient of determination is explained by the examples.

REFERENCES
1. https://en.oxforddictionaries.com/definition/correlation
2. "Definition of Statistics". www.merriam-webster.com. Retrieved 2016-
05-28. Web on 3 Nov 2017 4.30 pm in Wikipedia.
3. "Essay on Statistics: Meaning and Definition of Statistics". Economics
Discussion. 2014-12-02. Retrieved 2016-05-28. Web on 3 Nov 2017
4.30 pm in Wikipedia.
4. S.P. Gupta, “Statistical Methods.” Sultan Chand and Sons, Thirteenth
Revised Edition.2001.
5. M.G. Dhaygude,” Statistical and Quantitative Methods” Everest
Publishing House2004.Print.

Correlation Analysis 57
Statistical Techniques EXERCISE
1. From the data given below calculate the coefficient of correlation
between marks in mathematics and marks in statistics.
NOTES

2. From the data given below calculate the coefficient of correlation


between X and Y.

3. From the data given below calculate the coefficient of correlation,


coefficient of determination and probable error.

4. From the data given below calculate the coefficient of correlation,


coefficient of determination and probable error.

*****

58 Correlation Analysis
MODULE - 4
Statistical Techniques

NOTES
REGRESSION ANALYSIS

CHAPTER OBJECTIVES
1. To Distinguish between a dependent variable and an independent
variable and analyse data.
2. To Examine possible relationships between two variables.
3. To Develop simple linear regression models and use them as a
forecasting tool.

CONTENTS
4.1 Introduction
4.2 Definition of regression.
4.3 Regression lines.
4.4 Regression coefficient
4.5 Applications of regression in business.

KEY WORDS
Regression, Regression lines, Regression coefficient

OVERVIEW OF THE CHAPTER


This chapter focuses on the Regression lines. These lines are useful for
forecasting or estimating the value of one variable when the value of another
variable is known. It also includes the applications of regression in business.

4.1 INTRODUCTION

When there is a close relationship in between two variables we can estimate


the value of one variable when the value of another variable is known. Regression
Analysis is a Statistical tool used for estimating these unknown values of one
variable (say Dependent Variable) from the known values of the order (say
Independent Variable). This is accomplished through the regression line which
describes the relationship between two variables say X and Y. There are two types
of the regression simple regression and multiple regression.

Regression Analysis 59
Statistical Techniques

4.2 DEFINITION OF REGRESSION


NOTES
According to Wikipedia, Regression is a statistical measure used in finance,
investing and other disciplines that attempts to determine the strength of the
relationship between one dependent variable (usually denoted by Y) and a series
of other changing variables (known as independent variables).
According to business dictionary regression is a technique for determining
the statistical relationship between two or more variables where a change in a
dependent variable is associated with and depends on a change in one or more
independent variables.

4.3 REGRESSION LINES

The variables x and y have a linear relationship if a unit change in the


independent variable ( x or y ) results in a constant change in the dependent
variable ( y or x ). This is represented by an equation of a straight line of the
form, y = a + bx or x = a + by. There are two lines of regression,
Regression line y on x: This line of regression is used to estimate a value of
y for a given value of x and is expressed as below.

Regression line x on y: This line of regression is used to estimate a value of


x for a given value of y and is expressed as below.

4.4 REGRESSION COEFFICIENT

There are two regression coefficients. Regression coefficient y on x and


Regression coefficient x on y.
Regression coefficient y on x: (byx)

Regression coefficient x on y: (bxy)

60 Regression Analysis
Solutions: a. Statistical Techniques

NOTES

b. There appears to be a positive linear relationship between x and y.


c. Many different straight lines can be drawn to provide a linear
approximation of the relationship between x and y; in part (d) we will
determine the equation of a straight line that “best” represents the
relationship according to the least squares criterion.
d.

e.

2. a.

b. There appears to be a negative linear relationship between x and y.


c. Many different straight lines can be drawn to provide a linear
approximation of the relationship between x and y; in part (d) we will
determine the equation of a straight line that “best” represents the
relationship according to the least squares criterion.
Regression Analysis 61
Statistical Techniques d.

NOTES

e.

3. a.

b.

c.

4. a.

62 Regression Analysis
b. There appears to be a positive linear relationship between x = height Statistical Techniques
and y = weight.
c. Many different straight lines can be drawn to provide a linear
approximation of the relationship between x and y; in part (d) we will NOTES
determine the equation of a straight line that “best” represents the
relationship according to the least squares criterion.
d.

4.5 APPLICATIONS OF REGRESSION

Regression Analysis is widely used in various business sectors for


forecasting purpose. Below are the applications of regression models used in the
business are discussed.

1. Predictive Analytics
Predictive analytics i.e. forecasting future opportunities and risks is the most
prominent application of regression1 analysis in business. Demand analysis, for
instance, predicts the number of items which a consumer will probably purchase.
However, demand is not the only dependent variable when it comes to business.
Regression analysis can go far beyond forecasting impact on direct revenue. For
example, we can forecast the number of shoppers who will pass in front of a
billboard and use that data to estimate the maximum to bid for an advertisement.
Insurance companies heavily rely on regression analysis to estimate the credit
standing of policyholders and a possible number of claims in each time.

2. Operation Efficiency
Regression models can also be used to optimize business processes. A
factory manager, for example, can create a statistical model to understand the
impact of oven temperature on the shelf life of the cookies baked in those ovens.
In a call center, we can analyze the relationship between wait times of callers
and number of complaints. Data-driven decision making eliminates guesswork,
hypothesis and corporate politics from decision making. This improves the
business performance by highlighting the areas that have the maximum impact
on the operational efficiency and revenues.
Regression Analysis 63
Statistical Techniques 3. Supporting Decisions
Businesses today are overloaded with data on finances, operations and
customer purchases. Increasingly, executives are now leaning on data analytics
NOTES to make informed business decisions thus eliminating the intuition and gut feel.
Regression analysis can bring a scientific angle to the management of any
businesses. By reducing the tremendous amount of raw data into actionable
information, regression analysis leads the way to smarter and more accurate
decisions. This does not mean that regression analysis is an end to managers
creative thinking. This technique acts as a perfect tool to test a hypothesis before
diving into execution.

4. Correcting Errors
Regression is not only great for lending empirical support to management
decisions but also for identifying errors in judgment. For example, a retail store
manager may believe that extending shopping hours will greatly increase sales.
Regression analysis, however, may indicate that the increase in revenue might
not be sufficient to support the rise in operating expenses due to longer working
hours (such as additional employee labor charges). Hence, regression analysis
can provide quantitative support for decisions and prevent mistakes due to
manager's intuitions.

5. New Insights
Over time businesses have gathered a large volume of unorganized data that
has the potentialto yield valuable insights. However, this data is useless without
proper analysis. Regression analysis techniques can find a relationship between
different variables by uncovering patterns that were previously unnoticed. For
example, analysis of data from point of sales systems and purchase accounts may
highlight market patterns like increase in demand on certain days of the week or
at certain times of the year. You can maintain optimal stock and personnel before
a spike in demand arises by acknowledging these insights.

SUMMARY
In this chapter, we have studied the definitions of Regression, regression
lines. This chapter also focused on regression coefficients. In this chapter
applications of regression are also discussed.

REFERENCES
1. https://en.oxforddictionaries.com/definition/correlation
2. https://www.newgenapps.com
3. "Definition of Statistics". www.merriam-webster.com. Retrieved 2016-
05-28. Web on 3 Nov 2017 4.30 pm in Wikipedia.
4. "Essay on Statistics: Meaning and Definition of Statistics". Economics
Discussion. 2014-12-02. Retrieved 2016-05-28. Web on 3 Nov 2017
4.30 pm in Wikipedia.
5. S.P. Gupta, “Statistical Methods.” Sultan Chand and Sons, Thirteenth
Revised Edition.2001.
6. M.G. Dhaygude,” Statistical and Quantitative Methods” Everest
64 Regression Analysis Publishing House2004.Print.
EXERCISE Statistical Techniques
1. From the data given below calculate two regression lines and estimate
the marks of mathematics when marks of statistics are 50.
NOTES

2. From the data given below calculate two regression lines and
(i) Estimate X when Y is 58
(ii) Estimate Y when X is 30

3. From the data given below calculate two regression lines

4. From the data given below calculate two regression lines and
(i) Estimate X when Y is 70
(ii) Estimate Y when X is 60

5. Given:

Correlation Coefficient = r = 0.8 then Obtain two lines of regression.


6. Given the following information about advertising and sales.

Advertising Expenditure (in Rs lakhs) Sales (in Rs lakhs)


Correlation Coefficient = r = 0.8 then Obtain two lines of regression and
also find the likely sales when advertising expenditure is Rs 15 lakhs.
*****

Regression Analysis 65
MODULE - 5
Statistical Techniques

NOTES
PROBABILITY AND PROBABILITY
DISTRIBUTION

CHAPTER OBJECTIVES
1. Develop problem-solving techniques needed to accurately calculate
probabilities.
2. Apply problem-solving techniques to solving real-world events.
3. Apply selected probability distributions to solve problems.

CONTENTS
5.1 Sample space, events
5.2 Experiment, random experiment
5.3 Probability
5.4 Random variable
5.5 Conditional probability
5.6 Baye’s theorem
5.7 Probability distributions- binomial distribution, poisson distribution,
normal distribution

KEY WORDS
Sample space, Events, Random variable, Conditional probability, Baye’s
theorem, Binomial, Poisson, Normal distribution.

OVERVIEW OF THE CHAPTER


This chapter focuses on the sample space, events, experiment, random
experiment, random variable and probability. It also includes conditional
probability, Baye’s theorem and different probability distributions such as
Binomial distribution, Poisson distribution, and Normal distribution.

INTRODUCTION
Probabilities are associated with experiments where the outcome is not
known in advance or cannot be predicted. For example, if you toss a coin, will
you obtain a head or tail? If you roll a die will obtain 1, 2, 3, 4, 5 or 6? Probability
Probability And measures and quantifies "how likely" an event, related to these types of
Probability experiment, will happen. The value of a probability is a number between 0 and
66 Distribution
1 inclusive. An event that cannot occur has a probability (of happening) equal to Statistical Techniques
0 and the probability of an event that is certain to occur has a probability equal
to 1.(see probability scale below).
NOTES

To quantify probabilities, we need to define the sample space of an


experiment and the events that may be associated with that experiment.

5.1 SAMPLE SPACE AND EVENT

SAMPLE SPACE: The sample space is the set of all possible outcomes in
an experiment. It is denoted by S.
Examples
When a coin is tossed, S = {H, T} where H = Head and T = Tail
When a dice is thrown, S = {1, 2, 3, 4, 5, 6}
When two coins are tossed, S = {HH, HT, TH, TT} where H = Head and T = Tail
If two dice are rolled, the sample space S is given by
S = { (1,1),(1,2),(1,3),(1,4),(1,5),(1,6)
(2,1),(2,2),(2,3),(2,4),(2,5),(2,6)
(3,1),(3,2),(3,3),(3,4),(3,5),(3,6)
(4,1),(4,2),(4,3),(4,4),(4,5),(4,6)
(5,1),(5,2),(5,3),(5,4),(5,5),(5,6)
(6,1),(6,2),(6,3),(6,4),(6,5),(6,6) }

EVENT
An event is some specific outcome of an experiment. Any subset of a
Sample Space is an event. Events are generally denoted by capital letters A, B,
C, D etc.
Examples
When a coin is tossed, outcome of getting head or tail is an event
When a die is rolled, outcome of getting 1 or 2 or 3 or 4 or 5 or 6 is an event
A die is rolled. Let us define event E as the set of possible outcomes where Probability And
the number on the face of the die is even. Event E is given by E = {2,4,6} Probability
Distribution 67
Statistical Techniques Two coins are tossed (see example 2 above for the sample space). Let us
define event E as the set of possible outcomes where the number of head obtained
is equal to two. Event E is given by
NOTES E = {(HT), (TH)}
Two dice are rolled (see example 3 above for the sample space). Let us
define event E as the set of possible outcomes where the sum of the numbers on
the faces of the two dice is equal to four. Event E is given by E = {(1,3),
(2,2), (3,1)}
Events can be classified in various ways depending on different
characteristics.
Simple Events: Simple events are those events where only a single
experiment is carried out. Tossing of a coin or rolling a dice are known as simple
events. Simple events can be complementary to each other and independent or
dependent on each other. In the case of simple events, we take the probability of
occurrence of single events.

Examples
Probability of getting a Head (H) when a coin is tossed
Probability of getting 1 when a die is thrown
Compound Events: Compound events are those events which have more
than one experiments occurring together. For example, rolling a dice and tossing
a coin together will be known as a compound event. The sample space of
compound events is obtained using lists, table and tree diagrams. In the case of
compound events, we take the probability of joint occurrence of two or more
events.
Example: When two coins are tossed, probability of getting a Head (H) in
the first toss and getting a Tail (T) in the second toss.
Independent Events: Events can be said to be independent if the
occurrence or non-occurrence of one event does not influence the occurrence or
non-occurrence of the other. Independent events are those events where the
occurrence of one event will not affect the probability of occurrence of the other
event. If A and B are two independent events, then the probability of both A and
B will be written as

Example
If we toss a coin twice then the outcomes are independent of each other.
When a coin is tossed twice, the event of getting Tail(T) in the first toss and the
event of getting Tail(T) in the second toss are independent events. This is because
the occurrence of getting Tail(T) in any toss does not influence the occurrence
of getting Tail(T) in the other toss.
Dependent Events: Dependent events are those where occurrence of one
Probability And event will affect the other event.
Probability
68 Distribution
Example Statistical Techniques
If there are 3 red and 2 green balls in a bag and one green ball has been
taken out then the probability of getting a green ball in next attempt will get
affected. NOTES
Impossible Event: If the probability of an event is zero, then it will be
known as an impossible event. The empty set is known as the impossible event.
If the probability of an event is 1 then that event will occur. Such event is known
as sure event.
Exhaustive Event: Exhaustive Event is the total number of all possible
outcomes of an experiment. Events which together exhaust the whole sample
space are known as exhaustive events.

Example
Sample space = {1,2,3,4,5,6}, Odd = {1, 3, 5}, Even = {2, 4, 6}
The union of odd and even events will add up to the sample space.
When a coin is tossed, we get either Head or Tail. Hence there are 2
exhaustive events.
When two coins are tossed, the possible outcomes are (H, H), (H, T), (T,
H), (T, T). Hence there are 4 (=22) exhaustive events.
When a dice is thrown, we get 1 or 2 or 3 or 4 or 5 or 6. Hence there are 6
exhaustive events.
when we roll a die one event is to get all odd numbers and other is to get all
even numbers. Both the events together will exhaust the whole sample space.
Equally Likely Events: Events are said to be equally likely if there is no
preference for a event over the other.

Examples
When a coin is tossed, Head (H) or Tail is equally likely to occur.
When a dice is thrown, all the six faces (1, 2, 3, 4, 5, 6) are equally likely
to occur.
Mutually Exclusive Events: Two or more than two events are said to be
mutually exclusive if the occurrence of one of the events excludes the occurrence
of the other.

Examples
When a coin is tossed, we get either Head or Tail. Head and Tail cannot
come simultaneously. Hence occurrence of Head and Tail are mutually
exclusive events.
When a die is rolled, we get 1 or 2 or 3 or 4 or 5 or 6. All these faces cannot
come simultaneously. Hence occurrences of faces when rolling a die are mutually
exclusive events.
Probability And
Probability
Distribution 69
Statistical Techniques Note: If A and B are mutually exclusive events, A ∩∩ B = where
represents empty set.
Consider a die is thrown and A be the event of getting 2 or 4 or 6 and B be
NOTES the event of getting 4 or 5 or 6. Then A = {2, 4, 6} and B = {4, 5, 6}
Here A ∩ B ≠ . Hence A and B are not mutually exclusive events.

ALGEBRA OF EVENTS
Let A and B are two events with sample space S. Then
A U B is the event that either A or B or Both occur. (i.e., at least one of A
or B occurs)
A ∩ B is the event that both A and B occur
is the event that A does not occur
A∩ B is the event that none of A and B occurs
Example: Consider a die is thrown, A be the event of getting 2 or 4 or 6 and
B be the event of getting 4 or 5 or 6. Then
A = {2, 4, 6} and B = {4, 5, 6}, A B = {2, 4, 5, 6}, A ∩ B = {4, 6}, A¯
= {1, 3, 5}, B¯ = {1, 2, 3}, ¯A∩¯B = {1,3}

5.2 EXPERIMENT

An experiment is an operation which can produce well-defined outcomes.


The process of observing something uncertain is known as a random experiment.
This is the most basic term used in probability and statistics. The result of a
random experiment is an outcome and set of outcomes is sample space. The
probability of an event can be calculated using favorable outcomes and sample
space. The more number of repetitions are there for an experiment the bigger the
size of experiment is. The probability of each output of a random experiment can
be calculated if the sample space is known.
Random Experiment: If all the possible outcomes of an experiment are
known but the exact output cannot be predicted in advance, that experiment is
called a random experiment. Random experiment is the process to observe the
event having an uncertain outcome. When we toss a coin, the outcome is
uncertain and hence, it can be termed as a random experiment. The result of a
random experiment is known as the outcome and the set of all the possible
outcomes of an experiment is known as sample space. If we repeat an experiment
n number of times, then each time the experiment is done is known as a trial.
Random Experiment in Statistics: In statistics, the random experiment is
the observation of events having uncertain outcomes. For example, in bag ten
bulbs are there among which few are defective. If you start taking out one bulb
Probability And at a time and check if it is defective or not, this will be known as a trial.
Probability Observations are made from each trial, and in the end a statistical inference is
70 Distribution
made from it. The number of repetitions of experiment is known as size of Statistical Techniques
experiment. The statistical inference depends a lot on the size of experiment.

Examples
NOTES
Tossing of a fair coin: When we toss a coin, the outcome will be either Head
(H) or Tail (T)
Throwing an unbiased die: Die is a small cube used in games. It has six
faces and each of the six faces shows a different number of dots from 1 to 6.
Plural of die is dice.
When a die is thrown or rolled, the outcome is the number that appears on
its upper face and it is a random integer from one to six, each value being
equally likely.
Drawing a card from a pack of shuffled cards: A pack or deck of playing
cards has 52 cards which are divided into four categories as given below
Spades (♠) Clubs (♣) Hearts (♥) Diamonds (♦)
Each of the above-mentioned categories has 13 cards, 9 cards numbered
from 2 to 10, an Ace, a King, a Queen and a jack
Hearts and Diamonds are red faced cards whereas Spades and Clubs are
black faced cards.
Kings, Queens and Jacks are called face cards
Taking a ball randomly from a bag containing balls of different colours

5.3 PROBABILITY

Probability or chance is a common term used in day-to-day life. For


example, we generally say, 'it may rain today'. This statement has a certain
uncertainty. Probability is quantitative measure of the chance of occurrence of
an event.
Probability of an Event: Let E be an event and S be the sample space. Then
probability of the event E can be defined as

where, P(E) = Probability of the event E,


n(E) = number of ways in which the event can occur and
n(S) = Total number of outcomes possible
* Note : P(S) = 1
0 ≤ P (E) ≤ 1
P( ) = 0 ( Probability of occurrence of an impossible event = 0) Probability And
Probability
Distribution 71
Statistical Techniques Examples
A coin is tossed once. What is the probability of getting Head?
Total number of outcomes possible when a coin is tossed = n(S) = 2 (
NOTES Head or Tail)
E = event of getting Head = {H}. Hence n(E) = 1

Two dice are rolled. What is the probability that the sum on the top face of
both the dice will be greater than 9?
Total number of outcomes possible when a die is rolled = 6
(∵ any one face out of the 6 faces)
Hence, total number of outcomes possible two dice are rolled, n(S) = 6 × 6 = 36
E = Getting a sum greater than 9 when the two dice are rolled
= {(4, 6), {5, 5}, {5, 6}, {6, 4}, {6, 5}, (6, 6)}
Hence, n(E) = 6

GENERAL PROBABILITY
Probability is based on observations of certain events. Probability of an
event is the ratio of the number of observations of the event to the total numbers
of the observations. An experiment is a situation involving chance or probability
that leads to results called outcomes. An outcome is the result of a single trial of
an experiment. The probability of an event is the measure of the chance that the
event will occur because of an experiment.
Probability of an event A is symbolized by P(A).
Probability of an event A is lies between 0 ≤ P(A) ≤ 1.

PROBABILITY FORMULA
Probability is the measure of how likely an event is. And an event is one or
more outcomes of an experiment. Probability formula is the ratio of number of
favorable outcomes to the total number of possible outcomes.

Measures the likelihood of an event in the following way:


If P(A) > P(B) then event A is more likely to occur than event B.
If P(A) = P(B) then events A and B are equally likely to occur.
Probability And
Probability
72 Distribution
Basic rules of probability Statistical Techniques
Probability of an event lies between 0 and 1. If an event is certain to occur
it will have a probability of 1, and if there is no chance of the occurrence of an
event then the probability will be 0. Depending on various types of events, there NOTES
are different rules of probability. For example, if A and B are two independent
events, then the probability of A and B happening together will be product of
their individual probabilities. Again, if A and B are mutually exclusive events
then the probability of A and B happening together is zero.
These are the rules that are very important.
All probabilities are between 0 and 1 inclusive 0 <= P(E) <= 1
The sum of all the probabilities in the sample space is 1
The probability of an event which cannot occur is 0.
The probability of any event which is not in the sample space is zero.
The probability of an event which must occur is 1.
The probability of the sample space is 1.
The probability of an event not occurring is one minus the probability of
it occurring.
P(E') = 1 - P(E)

Definition
For any two events A and B, they can be classified as exclusive, exhaustive,
independent and dependent events. Based on the types of events they are, the
probability of A, P(A) and the probability of B, P(B) will have different rules of
probability for their addition, subtraction and multiplication.

Probability And
Probability
Distribution 73
Statistical Techniques

NOTES
For example: If in a biased coin the probability of getting a head is 0.8 then
what will be the probability of getting a tail?

Solution
The probability of getting a tail will be, 1−0.8=0.2.
Independent Probability Rules: Two events are known to be independent
events if occurrence of one event will have no effect on the probability of the
other one. If A and B are two independent events, and P(A) and P(B) be their
probabilities then independent probability rule of multiplication is,
Probability of both A and B occurring,

Example
Event of getting a king from the deck of 52 cards and getting a 6 if a dice
is rolled are independent events. Find the probability of both these events
happening together.

Solution
Probability of getting a king,
Probability of getting a 6,
Probability of both the events happening =

5.4 RANDOM VARIABLE

A random variable, usually written X, is a variable whose possible values


are numerical outcomes of a random phenomenon.
There are two types of random variables, discrete and continuous.

Discrete Random Variables


A discrete random variable is one which may take on only a countable
number of distinct values such as 0,1,2,3,4, ........ Discrete random variables are
usually (but not necessarily) counts. If a random variable can take only a finite
number of distinct values, then it must be discrete. Examples of discrete random
variables include the number of children in a family, the Friday night attendance
at a cinema, the number of patients in a doctor's surgery, the number of defective
light bulbs in a box of ten.

Probability And The probability distribution of a discrete random variable is a list of


Probability probabilities associated with each of its possible values. It is also sometimes
74 Distribution called the probability function or the probability mass function.
Suppose a random variable X may take k different values, with the Statistical Techniques
probability that X = xi defined to be P (X = xi) = pi. The probabilities pi must
satisfy the following:
0 < pi < 1 for each i NOTES
2: p1 + p2 + ... + pk = 1.

Example
Suppose a variable X can take the values 1, 2, 3, or 4.
The probabilities associated with each outcome are described by the
following table:
Outcome 1 2 3 4
Probability 0.1 0.3 0.4 0.2
The probability that X is equal to 2 or 3 is the sum of the two probabilities:
P(X = 2 or X = 3) = P(X = 2) + P(X = 3) = 0.3 + 0.4 = 0.7.
Similarly, the probability that X is greater than 1 is equal to 1 - P(X = 1) =
1 - 0.1 = 0.9, by the complement rule.
This distribution may also be described by the probability histogram shown
to the right:

All random variables (discrete and continuous) have a cumulative


distribution function. It is a function giving the probability that the random
variable X is less than or equal to x, for every value x. For a discrete random
variable, the cumulative distribution function is found by summing up the
probabilities.

Example
The cumulative distribution function for the above probability distribution
is calculated as follows:
The probability that X is less than or equal to 1 is 0.1,
The probability that X is less than or equal to 2 is 0.1+0.3 = 0.4,
The probability that X is less than or equal to 3 is 0.1+0.3+0.4 = 0.8, and Probability And
The probability that X is less than or equal to 4 is 0.1+0.3+0.4+0.2 = 1. Probability
Distribution 75
Statistical Techniques The probability histogram for the cumulative distribution of this random
variable is shown to the right:

NOTES

Continuous Random Variables: A continuous random variable is one


which takes an infinite number of possible values. Continuous random variables
are usually measurements. Examples include height, weight, the amount of sugar
in an orange, the time required to run a mile.
A continuous random variable is not defined at specific values. Instead, it
is defined over an interval of values, and is represented by the area under a curve
(in advanced mathematics, this is known as an integral). The probability of
observing any single value is equal to 0, since the number of values which may
be assumed by the random variable is infinite.
Suppose a random variable X may take all values over an interval of real
numbers. Then the probability that X is in the set of outcomes A, P(A), is defined
to be the area above A and under a curve. The curve, which represents a function
p(x), must satisfy the following:
• The curve has no negative values (p(x) > 0 for all x)
• The total area under the curve is equal to 1.
A curve meeting these requirements is known as a density curve.

5.5 CONDITIONAL PROBABILITY

Imagine a student who takes leave from school twice a week excluding
Sunday. If it is known that he will be absent from school on Tuesday, then what
are the chances that he will also take a leave on Saturday in the same week?
It is observed that in problems where the occurrence of one event affects
the happening of the following event. These cases of probability are known as
conditional probability.
Let us now investigate conditional probability definition mathematically,

Definition
Probability And The probability of occurrence of any event A when another event B in
Probability relation to A has already occurred is known as conditional probability. It is
76 Distribution depicted by P(A|B).
Statistical Techniques

NOTES

As depicted by above diagram, sample space is given by S and there are


two events A and B. In a situation where event B has already occurred, then our
sample space S naturally gets reduced to B because now the chances of
occurrence of an event will lie inside B.
As we must figure out the chances of occurrence of event A, only portion
common to both A and B is enough to represent the probability of occurrence of
A, when B has already occurred. Common portion of the events is depicted by
intersection of both the events A and B i.e. A ∩ B
This explains the concept of conditional probability problems i.e.
occurrence of any event when another event in relation to has already occurred.
Mathematically this can be represented as,

P(A \B) represents the probability of occurrence of A given B has occurred.


n(A ∩ B) is the number of elements common to both A and B.
n(B) is the number of elements in B and it cannot be equal to zero.
Similarly the probability of occurrence of B when A has already occurred
is given by,

To have a better insight let us practice some conditional probability


examples.

Example
Two die are thrown simultaneously, and the sum of the numbers obtained
is found to be 7. What is the probability that the number 3 has appeared at
least once?

Solution
The sample space S would consist of all the numbers possible by the
combination of two dies. Probability And
Probability
Therefore, S consists of 6 × 6 i.e. 36 events. Distribution 77
Statistical Techniques Event A indicates the combination in which 3 has appeared at least once.
Event B indicates the combination of the numbers which sum up to 7.
A= (3,1), (3,2), (3,3) (3,4) (3,5) (3,6) (1,3) (2,3) (4,3) (5,3) (6,3)
NOTES
B = (1,6) (2,5) (3,4) (4,3) (5,2) (6,1)

Applying the conditional probability formula, we get,

Example
Out of 50 people surveyed in a study, 35 runs in which there are 20 males.
What is the probability the if the person surveyed is a running then he is a male?

Solution
Probability of the person being male and a runner, P (A ∩ B) = 20/50
Probability of person being runner, P(A) = 35/50
Probability of a person being male if he is runner,

Hence, the compound probability that if the person surveyed is a runner


then he will be a male =

Example
The probability of raining on Sunday is 0.07. If today is Sunday, then find
the probability of rain today.

Solution
Probability that it is raining and the day is Sunday, P(A B)=0.07
Probability that is is Sunday, P(B) =
Probability that it will rain if today is Sunday, P(A|B) = = 0.49
Hence, the compound probability of raining if it is Sunday is 0.49.

Example
In a school the third language must be chosen between Hindi and French.
If a student has taken French then what is the probability that he will take Hindi,
if the probability of taking Hindi is 0.34?
Probability And
Probability
78 Distribution
Solution Statistical Techniques
Probability of taking French and Hindi, P (A ∩ B) =0 as they are mutually
exclusive events.
Probability of taking French, P(B)=0.34 NOTES

Probability of taking Hindi if French has been opted,

Compound probability of mutually exclusive events is 0.

5.6 BAYES’ THEOREM

Bayes’ theorem is a way to figure out conditional probability. Conditional


probability is the probability of an event happening, given that it has some
relationship to one or more other events. For example, your probability of getting
a parking space is connected to the time of day you park, where you park, and
what conventions are going on at any time. Bayes’ theorem is slightly more
nuanced. In a nutshell, it gives you the actual probability of an event given
information about tests.
“Events” Are different from “tests.” For example, there is a test for liver
disease, but that’s separate from the event of having liver disease.
Tests are flawed: just because you have a positive test does not mean you
actually have the disease. Many tests have a high false positive rate. Rare events
tend to have higher false positive rates than more common events. We’re not just
talking about medical tests here. For example, spam filtering can have high false
positive rates. Bayes’ theorem takes the test results and calculates your real
probability that the test has identified the event.
Bayes’ Theorem is a simple formula used to calculate conditional
probability. The Theorem was named after English mathematician Thomas Bayes
(1701-1761). The formal definition for the rule

5.7 PROBABILITY DISTRIBUTION

In probability theory and statistics, a probability distribution is a


mathematical function that provides the probabilities of occurrence of different
possible outcomes in an experiment. In more technical terms, the probability Probability And
distribution is a description of a random phenomenon in terms of the probabilities Probability
Distribution 79
Statistical Techniques of events. For instance, if the random variable X is used to denote the outcome
of a coin toss ("the experiment"), then the probability distribution of X would
take the value 0.5 for X = heads, and 0.5 for X = tails (assuming the coin is fair).
NOTES Examples of random phenomena can include the results of an experiment or
survey.
A probability distribution is defined in terms of an underlying sample space,
which is the set of all possible outcomes of the random phenomenon being
observed. The sample space may be the set of real numbers or a higher-
dimensional vector space, or it may be a list of non-numerical values; for
example, the sample space of a coin flip would be {heads, tails}.
Probability distributions are generally divided into two classes. A discrete
probability distribution (applicable to the scenarios where the set of possible
outcomes is discrete, such as a coin toss or a roll of dice) can be encoded by a
discrete list of the probabilities of the outcomes, known as a probability mass
function. On the other hand, a continuous probability distribution (applicable to
the scenarios where the set of possible outcomes can take on values in a
continuous range (e.g. real numbers), such as the temperature on a given day) is
typically described by probability density functions (with the probability of any
individual outcome being 0). The normal distribution is a commonly encountered
continuous probability distribution. More complex experiments, such as those
involving stochastic processes defined in continuous time, may demand the use
of more general probability measures.
A probability distribution whose sample space is the set of real numbers is
called univariate, while a distribution whose sample space is a vector space is
called multivariate. A univariate distribution gives the probabilities of a single
random variable taking on various alternative values; a multivariate distribution
(a joint probability distribution) gives the probabilities of a random vector – a
list of two or more random variables – taking on various combinations of values.
Important and commonly encountered univariate probability distributions include
the binomial distribution, the hypergeometric distribution, and the normal
distribution. The multivariate normal distribution is a commonly encountered
multivariate distribution.

TYPES OF PROBABILITY DISTRIBUTIONS


There are many different classifications of probability distributions. Some
of them include the normal distribution, chi square distribution, binomial
distribution, and Poisson distribution. The different probability distributions serve
different purposes. The binomial distribution, for example, evaluates the
probability of an event occurring several times over a given number of trials and
given the event's probability in each trial. The usual example would use a fair
coin and figuring the probability of that coin coming up heads in ten
straight flips.
The most commonly used distribution is the normal distribution and it is
Probability And used frequently in finance, investing, science, and engineering. The normal
Probability distribution is fully characterized by its mean and standard deviation, meaning
80 Distribution
the distribution is not skewed and does exhibit kurtosis. This makes the Statistical Techniques
distribution symmetric and it is depicted as a bell-shaped curve when plotted.
Therefore, there are the following types of distribution:
• Bernoulli Distribution NOTES
• Uniform Distribution
• Binomial Distribution
• Normal Distribution
• Poisson Distribution
• Exponential Distribution

BINOMIAL DISTRIBUTION
Let’s get back to cricket. Suppose that you won the toss today and this
indicates a successful event. You toss again but you lost this time. If you win a
toss today, this does not necessitate that you will win the toss tomorrow. Let’s
assign a random variable, say X, to the number of times you won the toss. What
can be the possible value of X? It can be any number depending on the number
of times you tossed a coin.
There are only two possible outcomes. Head denoting success and tail
denoting failure. Therefore, probability of getting a head = 0.5 and the probability
of failure can be easily computed as: q = 1- p = 0.5.
A distribution where only two outcomes are possible, such as success or
failure, gain or loss, win or lose and where the probability of success and failure
is same for all the trials is called a Binomial Distribution.
The outcomes need not be equally likely. Remember the example of a fight
between me and Undertaker? So, if the probability of success in an experiment
is 0.2 then the probability of failure can be easily computed as q = 1 – 0.2 = 0.8.
Each trial is independent since the outcome of the previous toss doesn’t
determine or affect the outcome of the current toss. An experiment with only two
possible outcomes repeated n number of times is called binomial. The parameters
of a binomial distribution are n and p where n is the total number of trials and p
is the probability of success in each trial.
Based on the above explanation, the properties of a Binomial Distribution are
Each trial is independent.
There are only two possible outcomes in a trial- either a success or a failure.
A total number of n identical trials are conducted.
The probability of success and failure is same for all trials. (Trials are
identical.)
The mathematical representation of binomial distribution is given by:

Probability And
A binomial distribution graph where the probability of success does not Probability
equal the probability of failure looks like Distribution 81
Statistical Techniques

NOTES

Now, when probability of success = probability of failure, in such a situation


the graph of binomial distribution looks like

The mean and variance of a binomial distribution are given by:


Mean -> µ = n*p
Variance -> Var(X) = n*p*q

NORMAL DISTRIBUTION
Normal distribution represents the behavior of most of the situations in the
universe (That is why it’s called a “normal” distribution. I guess!). The large sum
of (small) random variables often turns out to be normally distributed,
contributing to its widespread application. Any distribution is known as Normal
distribution if it has the following characteristics:
The mean, median and mode of the distribution coincide.
The curve of the distribution is bell-shaped and symmetrical about the line x=μ.
The total area under the curve is 1.
Exactly half of the values are to the left of the center and the other half to
the right.
A normal distribution is highly different from Binomial Distribution.
However, if the number of trials approaches infinity then the shapes will be quite
similar.
The PDF of a random variable X following a normal distribution is given by:

Probability And
Probability
82 Distribution
The mean and variance of a random variable X which is said to be normally Statistical Techniques
distributed is given by:
Mean -> E(X) = µ
Variance -> Var(X) = σ^2 NOTES

Here, µ (mean) and σ (standard deviation) are the parameters.


The graph of a random variable X ~ N (µ, σ) is shown below.

A standard normal distribution is defined as the distribution with mean 0


and standard deviation 1. For such a case, the PDF becomes:

POISSON DISTRIBUTION
Suppose you work at a call center, approximately how many calls do you
get in a day? It can be any number. Now, the entire number of calls at a call center
in a day is modeled by Poisson distribution. Some more examples are
The number of emergency calls recorded at a hospital in a day.
The number of thefts reported in an area on a day.
The number of customers arriving at a salon in an hour.
The number of suicides reported in a particular city.
The number of printing errors at each page of the book. Probability And
You can now think of many examples following the same course. Poisson Probability
Distribution 83
Statistical Techniques Distribution is applicable in situations where events occur at random points of
time and space wherein our interest lies only in the number of occurrences of
the event.
NOTES A distribution is called Poisson distribution when the following assumptions
are valid:
1. Any successful event should not influence the outcome of another
successful event.
2. The probability of success over a short interval must equal the
probability of success over a longer interval.
3. The probability of success in an interval approaches zero as the interval
becomes smaller.
Now, if any distribution validates the above assumptions then it is a Poisson
distribution. Some notations used in Poisson distribution are:
λ is the rate at which an event occurs,
t is the length of a time interval,
And X is the number of events in that time interval.
Here, X is called a Poisson Random Variable and the probability distribution
of X is called Poisson distribution.
Let µ denote the mean number of events in an interval of length t. Then, µ = λ*t.
The PMF of X following a Poisson distribution is given by:

The mean µ is the parameter of this distribution. µ is also defined as the λ


times length of that interval. The graph of a Poisson distribution is shown below:

The graph shown below illustrates the shift in the curve due to increase in mean.

Probability And
Probability
84 Distribution
It is perceptible that as the mean increases, the curve shifts to the right. Statistical Techniques
The mean and variance of X following a Poisson distribution:
Mean -> E(X) = µ
NOTES
Variance -> Var(X) = µ
Relation between Poisson and Binomial Distribution
Poisson Distribution is a limiting case of binomial distribution under the
following conditions:
The number of trials is indefinitely large or n → ∞.
The probability of success for each trial is same and indefinitely small or p →0.
np = λ, is finite.
Relation between Normal and Binomial Distribution & Normal and Poisson
Distribution:
Normal distribution is another limiting form of binomial distribution under
the following conditions:
The number of trials is indefinitely large, n → ∞.
Both p and q are not indefinitely small.
The normal distribution is also a limiting case of Poisson distribution with
the parameter λ and ∞.

Probability Distribution Examples


The likelihood of an event to occur is known as its probability. The
probability distribution is the distribution table formed to obtain the occurrence
of each event in the sample space. From the probability distribution table, we
can find the mean, standard deviation, and expected value.
Example: 1
If a coin is tossed, find the probability distribution of getting tails.

Solution
If a coin is tossed, either 0 tail or 1 tail can be obtained.
Probability of getting zero tail, P(0) = 1/2
Probability of getting one tail, P(1) = 1/2
The probability distribution table will be,

Example 2
If two coins are tossed simultaneously, find the probability distribution of Probability And
getting head. Probability
Distribution 85
Statistical Techniques Solution
If two coins are being tossed, there can be either 0 heads, 1 head, or 2 heads.
Sample space, S = {HH, HT, TH, TT}
NOTES
The probability distribution of getting heads can be shown as:

Example 3
In a class on 100 students, 80 students passed in all subjects, 10 failed in
one subject, 7 failed in two subjects and 3 failed in three subjects. Find the
probability distribution of the variable for number of subjects a student from the
given class has failed in.

Solution
For a random student,
The probability of failing in 0 subjects, P(X=0) = 0.8
The probability of failing in 1 subjects, P(X=1) = 0.1
The probability of failing in 2 subjects, P(X=2) = 0.07
The probability of failing in 3 subjects, P(X=3) = 0.03
The probability distribution can be shown as:

Example 4
A die is having 6 sides has two dots on 3 sides, four dots on 2 sides and six
dot on 1 side. Find the probability distribution of getting a number on rolling
the die.

Solution
On rolling the given die, any one of the three numbers, 2, 4, and 6 will come.
Probability of getting 2 dots, P(X = 2) = 3/6 = 1/2
Probability of getting 4 dots, P(X = 4) = 2/6 = 1/3
Probability of getting 6 dots, P(X = 6) = 1/6
The probability distribution can be shown as,

Probability And
Probability
86 Distribution
Example 5 Statistical Techniques
Three coins are being tossed, find the probability distribution of getting any
number of tails.
NOTES
Solution
Sample space, S = {HHH, HHT, HTH, THH, HTT, THT, TTH, TTT}
Hence, in tossing three coins number of tails can be 0, 1, 2, 3.
Probability of getting 0 tails, P(X=0) = 1/8
Probability of getting 1 tails, P(X=1) = 3/8
Probability of getting 2 tails, P(X=2) = 3/8
Probability of getting 3 tails, P(X=3) = 1/8
Probability distribution can be given as:

Binomial Distribution Examples


Binomial distribution is applied on the experiments which can give only
two outcomes. In binomial probability, if probability of success is pp, then
probability of failure is (1−p). In a binomial experiment, for xx number of
successes in n trials, where pp is the probability of success and q is the probability
of failure can be given as: P(x) = n C x .Px qn - x
To find the probability of number of success less than a given value x, the
binomial distribution can be given by,
P(X= n)= P(X=0)+P(X=1)+...+P(X=n)P(X=n)=P(X=0)+P(X=1)+...+P(X=n)

Example 6
The probability of winning a match for team AA is 0.60.6. Find the
probability of winning 33 matches out of 55.

Solution
Probability of winning, pp = 0.60.6
Probability of losing, q = 0.40.4
Probability of winning 33 matches out of 5, i.e P(x =3)
P(x = 3) = 5!3!2!5!3!2! × 0.216× 0.16
=0.3456× 0.216× 0.16
=0.3456
Hence, the probability is 0.3456
Probability And
Probability
Distribution 87
Statistical Techniques Example 7
If a committee has 77 members, find the probability of having more female
members than male members given that the probability of having a male or a
NOTES female member is equal.

Solution
The probability of having a female member = 0.50
The probability of having a male member = 0.50
To have more female members, the number of females should be greater
than or equal to 44.
P(X ≥ 4) = P(4)+P(5)+P(6)+P(7)P(4)+P(5)+P(6)+P(7) = C74(0.5)4 (0.5)
3+C75(0.5)5(0.5)2+C76(0.5)6(0.5)1+C77(0.5)7(0.5)0C47(0.5)
4(0.5)3+C57(0.5) 5(0.5)2+C67(0.5)6(0.5)1+C77(0.5)7(0.5)0
= (0.5)7×(C74+C75+C76+C77)
=0.0078125×64
=0.5(0.5)7×(C47+C57+C67+C77)
=0.0078125×64=0.5
The probability is 0.5

Example 8
Aren is taking part in four competitions. If the probability of him winning
any competition is 0.30.3, find the probability of him winning at least one
competition.

Solution
Probability of winning at least one competition will be the complement of
the probability of winning not a single competition.
P(X=0) = 40C(0.3)0(0.7)404C(0.3)0(0.7)4
= 1×1×0.2401
= 0.24011×1×0.240
= 0.2401

Chances of winning at least one competition = 1−P(X=0)


=1−0.2401
= 0.75991−P(X=0)
= 1−0.2401
= 0.7599

Probability And
Probability
88 Distribution
Example 9 Statistical Techniques
If a coin is tossed thrice, find the probability of a getting head at least two times.

Solution NOTES
The probability of getting head at least two times is the sum of probabilities
of getting head two times and three times.
P(X ≥ 2) = P(X=2)+P(X=3)
= 32C(0.5)2(0.5)1+33C(0.5)3(0.5)023C(0.5)2(0.5)1+33C(0.5)
3(0.5)0
= 3×0.125+1×0.125
= 0.53×0.125+1×0.125
= 0.5
Hence, the needed probability is 0.50.

Example 10
If only 55 percent kids can secure A grade in a paper, find the probability
of at most 22 out of 1010 kids getting A grade in that paper.

Solution
Probability of securing grade A=0.05
Probability of at most 22 kids getting grade A=P(X=0)+P(X=1)+P(X=2)
The required probability =
C100(0.05)0(0.95)10+C101(0.05)1(0.95)9+C102(0.05)2(0.95)8C010(0.05)
0(0.95)10+C110(0.05)1(0.95)9+C210(0.05)2(0.95)8
= 0.5987+ 0.3151 +0.0746
= 0.9884
Hence, the probability needed is 0.9884

Poisson Distribution Examples


The Poisson distribution is used when the mean of occurrence of a certain
event is given for a certain time and the probability is needed to be calculated
for a certain value in the same time. Let the mean be m, then the probability for
the event to occur x times in the given time will be,
P(x) = e-m mx / x!

Example 11
The mean value for an event X to occur is 2 in a day. Find the probability
of event X to occur thrice in a day.
Probability And
Probability
Distribution 89
Statistical Techniques Solution
Mean, m = 2
Probability of the event to occur thrice, P (3;2) = e−2 = 233! = 0.1804
NOTES
Example 12
A shop sells five pieces of shirt every day, then what is the probability of
selling three shirts today?

Solution
Mean value for 1 day, m = 5
Probability of selling 33 shirts, P(3;5)= e−5 = 533!
= 0.0067×1255!
= 0.0070
Hence, the probability of selling three shirts is 0.007 when at the average
55 shirts are being sold each day.

Example 13
If three persons, on an average, come to ABC company for job interview,
then find the probability that less than three people have come for interview on
a given day.

Solution
The mean for Poisson random variable, m=3
P(x<3;3) = P (0;3) + P (1;3) + P (2;3) P(x<3;3)
P (0;3) = e−3e−3 300!300!
= 0.0497
= 0.1493
P (2;3) = e−3e−3 322!322!
= 0.2240
Hence,
P(x <3;3) = P (0;3) + P (1;3) + P (2;3)
= 0.0497 +0.1493 +0.2240
= 0.4231
The probability of less than three persons coming for interview on a certain
day is 0.4231

Example 14
Number of calls coming to the customer care center of a mobile company
per minute is a Poisson random variable with mean 5. Find the probability that
Probability And
Probability no call comes in a certain minute.
90 Distribution
Solution Statistical Techniques
The mean value, m=5m=5
We need to find the probability of getting zero calls when 5 calls are known
to come every minute. NOTES

P(0;5) = e−5e−5 500!500!


= 0.0067379470
Hence, the probability of getting zero calls in a minute is 0.0067

Example 15
There are five students in a class and the number of students who will
participate in annual day every year is a Poisson random variable with mean 3.
What will be the probability of more than 3 students participating in annual day
this year?

Solution
Mean for Poisson random variable, m = 3
P(x>3;3) = P (4;3) + P (5;3)
P (4;3) = e−3e−3 344!344!
= 0.1680
P (5;3) = e−3e−3 355!355!
= 0.1008
Hence, P(x>3;3) = P (4;3) + P(5;3)
= 0.268850169
The probability of getting more than three students participating is 0.2688

Example 16
The deals cracked by an agent per day is a random Poisson variable with
mean 2. Given that each day is independent of other day, find the probability of
getting 2 deals cracked on first day and 1 deal to be cracked the next day.

Solution
The probability of getting 2 deals in a day is P(2;2) and the probability of
getting 1 deal is P(1;2).
The probability of getting 2 deals on first day and one deal on second day
= P(2;2) × P(1;2)
PP = e−2e−2 222!222!
= 0.2706 + 0.2706
= 0.5413
The probability the first day two deals are cracked, and the second day one Probability And
deal is cracked is 0.5413 Probability
Distribution 91
Statistical Techniques Normal Distribution Examples
Normal distribution is a symmetric distribution where the single peak is at
the mean of the data. The normal distribution curve is bell shaped and the spread
NOTES of data is controlled by the standard deviation. The 68-95-99.7 rule says that 68
percent of data in a normal distribution comes under one standard deviation, 95
percent comes under two standard deviations and 99.7 percent of data comes
under three standard deviations. To get the probability of from a normal
distribution, we need to know the mean and the standard deviation of the given
data. Then, calculating the z-score and looking up to the z- table gives the needed
probability.
Formula to find the z-score is,
Z=x−m/σ

Example 17
If mean of a given data for a random value is 81.1 and standard deviation
is 4.7, then find the probability of getting a value more than 83.

Solution
Standard deviation, σ = 4.7
Mean, m = 81.1
Expected value, X = 83
Z-score, Z = x− m / σ
z = 83−81.1 / 4.7 = 0.4042
Looking up the z-score in the z-table, 0.6700
Hence, probability (1−0.6700) = 0.330

Example 18
The average speed of a car is 65 kmph with a standard deviation of 4. Find
the probability that the speed is less than 60 kmph.

Solution
Mean m = 65
Standard deviation, σ = 4
Expected value, X = 4
Z-score, Z = x− m / σ
z = 60 − 65 / 4
= -1.25
Looking up the z-score in the z-table, we get 0.1056
Hence, probability is 0.1056.
Probability And
Probability
92 Distribution
Example 19 Statistical Techniques
The average score of a statistics test for a class is 85 and standard deviation
is 10. Find the probability of a random score falling between 75 and 95.
NOTES
Solution
The probability of score falling between 75 and 95 can be found after
finding the respective z-scores.
For X = 75, z = 75 – 85 / 10
= -1.001
For X = 95, z = 95−85 / 10
= 2.002
Probability is, P (-1.00 < z < 2.00) = P(z < 2.00) - P(z < -1.00)
= 0.9772 - 0.1587 = 0.8185.

SUMMARY
In this chapter, we have studied the general definition of probability, random
experiments. Conditional probability and Baye’s theorem is also studied. The
chapter discussed about probability distributions such as normal distribution,
Poisson distribution and Binomial distribution with different types of examples.

REFERENCES
1. Valerie J. Easton and John H. McColl's Statistics Glossary v1.1
2. https://en.wikipedia.org/wiki/Probability_distribution
3. https://www.analyticsvidhya.com/blog/2017/09/6-probability-
distributions-data-science/
4. https://www.investopedia.com/terms/p/probabilitydistribution.asp#i
xzz5HFYZkGKb
5. http://www.probabilityformula.org/probability-distribution-
examples.html
6. http://www.stat.yale.edu/Courses/1997-98/101/ranvar.htm
7. www.analyzemath.com

EXERCISE
1. 40% of people can roll their tongue, and 10% of people surveyed had
red hair.
a) What is the probability of someone having red hair and having
the ability to roll their tongue?
b) In a random sample of six people, what is the probability that at
least five of them can roll their tongue?
Probability And
Probability
Distribution 93
Statistical Techniques 2. A bridge hand is made up of 13 cards from a regular 52 card deck.
a) what is the probability that a player has three aces in his hand?
b) What is the probability that a player does not have a face card or
NOTES an ace in his hand?

3. Define the following terms and give an example of each:


a) Dependent events
b) Mutually exclusive events

4. A bag contains 8 white balls and 5 red balls.What is the probability of:
a) drawing 3 white balls with your first three attempts, without
replacement?
b) drawing 15 balls, replacing the ball each time and drawing out
exactly 9 red balls?

5. Five-person committee is to be formed from four teachers and four


students.
a) What is the probability that there is exactly one teacher on the
committee?
b) What is the probability that students outnumber teachers on the
committee?

6. What is the probability that in a group of 22 people at least two people


have the same birthday?

7. What is the probability that in a group of 7 people at least 2 people


share the same birth month?

8. A building supply store buys 40% of its pine boards from sawmill A
and 60% from sawmill B. 7% of the boards from sawmill A and 5
percent from sawmill B have blue discoloration. If a randomly picked
board is discolored, what is the probability that it came from
sawmill A?

9. A bag contains 4 red marbles and 3 blue marbles.


a) What is the probability of selecting a red marble?
b) What are the odds of selecting a red marble?

10. In tossing a coin 12 times, what is the probability of obtaining


a) at least 10 heads?
Probability And b) at least 2 heads?
Probability
94 Distribution
11. In a 5 card poker hand, what is the probability of having Statistical Techniques
a) exactly four spades in your hand?
b) no more than one red card in your hand?
NOTES
12. In a random sample of Tung Hai University students 50% indicated
they are business majors, 40% engineering majors, and 10% other
majors. Of the business majors, 60% were female; whereas, 30% of
engineering majors were females. Finally, 20% of the other majors
were female. Given that a person is female, what is the probability that
she is an engineering major?

13. The number of customers arriving at Taiwan Bank is Poisson


distributed with a mean, 4 customers/per minute.
(a) Within 2 minutes, what is the probability that there are 3
customers?
(b) What is the probability density function for the time between the
arrival of the next customer?

14. Suppose the average number of car accidents on the highway in one
day is 4. What is the probability of no car accident in one day? What
is the probability of 1 car accidence in two days?

15. Suppose the average number of calls by 104 in one minute is 2. What
is the probability of 10 calls in 5 minutes?

16. Replacement times for CD players are normally distributed with a


mean of 7.1 years and a standard deviation of 1.4 years (data from
Consumer Reports). What is the probability that a randomly-selected
CD player will have to be replaced in 8 years or less?

17. A mile-runner’s times for the mile are normally distributed with a
mean of 4 min. 3 sec. (This would have to be expressed in decimal
minutes -- 4.05 minutes), and a standard deviation of 2 seconds
(0.0333333••• minutes (the three dots indicate a repeating decimal)).
What is the probability that on a given run, the time will be 4 minutes
or less?

18. A machine fills 24-ounce (according to the label) boxes with cereal.
The amount deposited into the box is normally distributed with a
standard deviation of 0.25 ounce. What does the mean have to be for
99.5% of the boxes to contain 24 ounces or more of cereal?

*****
Probability And
Probability
Distribution 95
MODULE - 6
Statistical Techniques

NOTES
ASSOCIATION OF ATTRIBUTES

CHAPTER OBJECTIVES
1. To determine the association between two or more attributes.
2. To study Yules association of attributes.

CONTENTS
6.1 Introduction,
6.2 Types of association
6.3 Notations, order of classes and class frequencies, two attributes and
three attributes,
6.4 Yule’s coefficient of association and interpretation.
6.5 Decision making; process of decision making, types of decision, risk,
uncertainty, criteria of decision making

KEY WORDS
Association, Attributes, Risk, Uncertainty

OVERVIEW OF THE CHAPTER


This chapter focuses on association of attributes. It also focuses on
association between two and three attributes. This chapter includes Yule’s
coefficient of association method. This chapter focuses on decision making
process, types of decision and various criteria of decision making.

6.1 INTRODUCTION

The meaning of association in statistical language is quite different from


the common meaning of association. Commonly, if two attributes A and B appear
together number of times then they can be said to be as associated. But according
to Yule and Kendall, “In Statistics A and B are associated only if they appear
together in a greater number of cases than is to be expected, if they are
independent.”
Methods used to measure the association of attributes refer to those
techniques, which are used to measure the relationship between two such
Association of phenomena, whose size cannot be measured and where we can only find the
96 Attributes presence or absence of an attribute.
In the case of correlation analysis, we study the relationship between two Statistical Techniques
variables, which we can measure quantitatively. Similarly, in the case of
association we study the relationship between two attributes, which are not
quantitatively measurable. For example, level of education and crime. In
NOTES
association, no variables are involved. As it has been stated earlier an attribute
divides the universe into two classes, one possessing the attribute and another
not possessing the attribute whereas the variable can divide the universe into any
number of classes. Correlation coefficient is a measure of degree or extent of
linear relationship between two variables, whereas the coefficient of association
indicates association between two attributes and also whether the association is
positive or negative. But with the help of coefficient of association we cannot
find expected change in A for a given change in B and vice-versa, as possible by
regression coefficient, which is derived from correlation coefficient.

6.2 TYPES OF ASSOCIATION

Two attributes A and B are said to be associated if they are not independent
but are related in some way or the other. There are three kinds of associations,
which possibly occur between attributes.
1. Positive association
2. Negative association or disassociation
3. No association or independence.
In positive association, the presence of one attribute is accompanied by the
presence of another attribute. For example, health and hygiene are positively
associated.
or if (AB ) ( A )( B )
N
Then attributes A and B are positively associated.
In negative association, the presence of one attribute say A ensures the
absence of another attribute say B or vice versa. For example, vaccination and
occurrence of disease for which vaccine is meant are negatively associated.
or if (AB ) ( A )( B )
N
Then attributes A and B are negatively associated.
If two attributes are such that presence or absence of one attribute has
nothing to do with the absence or presence of another, they are said to
independent or not associated. For example, Honesty and Boldness
or if (AB ) ( A )( B )
N
Then attributes A and B are independent. Association of
Attributes 97
Statistical Techniques

6.3 NOTATIONS
NOTES
In this unit, following symbols will be used
(AB)0 = (A)(B) / N , (αβ)0 = (α)(β) / N
(αB)0 = (α)(B) / N , (Aβ)0 = (A)(β) / N
δ = (AB) – (AB)0 = (AB) – (A)(B) / N
If δ = 0 then (AB) = (A)(B) / N ……… A and B are independent
If δ > 0 then attributes A and B are positively associated and if < 0 then
attributes A and B are negatively associated.

Presentation of attributes in table form


If we have 2 attributes, we have 9 classes. We can represent the data in the
table form called as 2 X 2 contingency table or 9 square table as below

If we have 3 attributes, we have 27 classes. We can represent the data in the


table form called as 3 X 3 contingency table.

Example 1
Show whether A and B are independent, positively associated or negatively
associated in each of the following cases:
(i) N = 1000; (A) = 550; (B) = 700; (AB) = 540
(ii) (A) = 480; (AB) = 290; (α ) = 585; ( B) = 383
(iii) N = 1000; (A) = 500; (B) = 400; (AB) = 200

Solution
We have given
i) (A)(B) / N = (550) (700) / 1000 = 385 = (AB)0
Thus (AB) = 540 > (A)(B) / N
Since (AB) > (AB)0 hence they are positively associated.

(ii) (B) = ( AB ) + ( αB ) = 290 + 383 = 673


N = (A) (α) = (480) + (585) = 1065
(A)(B) / N = 480 * 673 / 1065 = 303.32 = (AB)0
Thus (AB) = 290 < 303.32

Association of Thus (AB) < (AB)0


98 Attributes Hence A and B are negatively associated.
(iii) (A)(B) / N = 500*400 / 1000 = 200 = (AB)0 Statistical Techniques
Thus, hare (AB) = (AB)0
Hence, A and B are independent, i.e. δ = 0
NOTES

6.4 YULE’S COEFFICIENT OF ASSOCIATION

G. Udny Yule has developed a formula for finding the association


between two variables.
For two attributes A and B, the coefficient of association is given as
Q = (AB)( αβ ) - ( Aβ )( αB ) / ( AB )( αβ ) + (Aβ)(ΑB)

Value of Q lies between –1 and +1.


1. If Q = 1, A and B has perfect positive association
2. If Q = –1, A and B possess perfect negative association.
3. If Q = 0, A and B are independent.

Example 2
In a certain interview, there were 150 candidates of which 80 were boys,
40 candidates were successful among them 15 were boys. Calculate Yule’s
Coefficient of Association.

Solution
Let A be the boys; α be the girls and B be the successful; β be failed
We have N = 150, (A) = 80, (B) = 40, (AB) = 15
We have prepared 2 x 2 contingency table as below

Now, Yule’s Coefficient of Association


Q = (AB)(αβ ) - ( Aβ )( αB ) / ( AB )( αβ ) + (Aβ)(ΑB)
= (15)(45) – (65)(25) / (15)(45) + (65)(25)
= (675) - (1625) / (675) + (1625)
= -950 / 2300
Q = -0.41

Association of
Attributes 99
Statistical Techniques Example 3
Calculate Yule’s coefficient of association for the following data:
(i) (A) = 500; (B) = 400 ;(AB) = 350; N = 1000
NOTES
(ii) (A) = 900; (B) = 800; (AB) = 600; N = 2000

Solution
(i) We have prepared 2 x 2 contingency table as below

Yule’s Coefficient of Association


Q = (AB)(αβ ) - ( Aβ )( αB ) / ( AB )( αβ ) + (Aβ)(ΑB)
= (350)(450) – (150)(50) / (350)(450) + (150)(50)
= (157500) - (7500) / (157500) + (7500)
= 150000 / 165000
Q = 0.90

ii) We have prepared 2 x 2 contingency table as below

Yule’s Coefficient of Association


Q = (AB)(αβ ) - ( Aβ )( αB ) / ( AB )( αβ ) + (Aβ)(ΑB)
= (600)(900) – (300)(200) / (600)(900) + (300)(200)
= (540000) - (60000) / (540000) + (60000)
= 480000 / 600000
Q = 0.8

Example 4
Investigate if there is any association between extravagance in father and
son from the following:
Extravagant sons with extravagant fathers (AB) = 600
Miser sons with extravagant fathers ( B) = 1750
Extravagant sons with miser fathers (A ) = 1000
Association of
Miser sons with miser fathers (αβ) = 350
100 Attributes
Solution Statistical Techniques
We have prepared 2 x 2 contingency table as below

NOTES

Yule’s Coefficient of Association


Q = (AB)(αβ ) - ( Aβ )( αB ) / ( AB )( αβ ) + (Aβ)(ΑB)
= (600)(350) – (400)(1150) / (600)(350) + (400)(1150)
= (210000) - (460000) / (210000) + (460000)
= -250000 / 670000
Q = 0.37

6.5 DECISION THEORY

When linear programming models are used, decisions are made based on
the results of the models. However, these LP models were all formulated under
the assumption that certainty exists. It is assumed that all the model coefficients,
constraint values, and solution values are known with certainty and do not vary.
Decisions are often made within an environment of risk, uncertainty, or conflict.
Decision theory is concerned with decision making under conditions of risk and
uncertainty. Game theory, which will not be discussed, is concerned with decision
making under conflict. Decision situations can be categorized into two classes:
situations where probabilities cannot be assigned to future occurrences, and
situations where probabilities can be assigned.

Steps of Decision Making


A decision-making situation includes a couple of steps. The first step in the
decision analysis is to identify the alternatives considered by the decision maker.
These alternatives are called decision alternatives, denoted by d1, d2, d3, ....
d1 = invest in the apartment building,
d2 = invest in the office building,
d3 = invest in the warehouse,
d4 = don't invest.
Of course, the selection of the best decision alternative depends on how
much profit the investor will make from each decision. Profit will depend on the
future events associated with a decision situation. A decision maker may have
an idea of possible future events, but it is uncertain which particular event will
occur. Thus the second step in the analysis is to identify the future events that Association of
Attributes 101
Statistical Techniques might occur. These events are referred to as the states of nature, denoted by s1,
s2, s3,... The states of nature are defined so that one of the listed states of nature
will occur.
NOTES s1 = good economic conditions
s2 = poor economic conditions.
Given the three decision alternatives and the two states of nature, which
investment opportunity should the decision maker choose? In order to answer
this question, we need information on the profit associated with each combination
of a decision alternative and a state of nature. For example, how much profit
would the investor experience if she/he decides to purchase the apartment
building (d1) and economic conditions are good (s1)? And so on. The outcome
resulting from making a certain decision and the occurrence of a particular state
of nature is referred to as payoff. We will organize these types of decision
situations into payoff tables. Entries in a payoff table can be stated in terms of
profits, costs, or any other measure of output that may be appropriate for the
particular situation. Below is an example:

Decision Making Without Probabilities


In this case, we consider decision-making approaches without knowledge
of the probabilities of the states of nature. Once a payoff table is organized,
several criteria are available for decision-making. They will each yield different
decisions. The decision maker must select the criterion or combination of criteria
that best suits his or her needs. The criteria are:
• The Maximax Criterion
• The Maximin Criterion
• The Minimax Regret Criterion
• The Hurwitz Criterion
• The Laplace Criterion
When a compound word like "maximax" is given, the first portion "max"
chooses the maximum among the maxima of the second portion "max." The
"maximin" means that we select the maximum among the minima.

The Maximax Criterion


The maximax criterion assumes of an optimistic decision maker. With this
criterion, the decision maker selects the alternative that represents the maximum
of the maximum payoff. For the above example, the maximum payoff for each
Association of of the four investment plans is highlighted:
102 Attributes
Statistical Techniques

NOTES

Then the maximum among the maximum values is 550,000, so the decision
is to construct a "large-sized facility."

The Maximin Criterion


The reverse approach to the maximax criterion is the maximin criterion,
which assumes that the decision maker is conservative (pessimistic) about the
future. With this criterion, the minimum payoffs for each alternative are
compared, and the alternative that produces the maximum of the minimum
payoffs must be chosen. In the above example, this would be:

So the decision maker would choose "No facility" because a payoff of 0 is


the maximum among the minimum payoffs. The maximin is designed for the
decision maker who wants to avoid risk.

The Minimax Regret Criterion


We feel regret if we did not get what we wanted, or we discovered that what
we chose produced less than expected. We define the regret function under each
state of nature as
As you see, there is no regret when you get the best payoff under each state
of nature. We construct a regret (opportunity loss) table:

Each of the highlighted payoffs is the maximum for each alternative. The
minimum among the highlighted maximum is 250,000 for a "medium-sized
facility." The decision should be to construct a medium-sized facility.

The Hurwitz Criterion


This criterion strikes a compromise between the maximax and maximin Association of
criteria. The decision maker is neither optimistic nor pessimistic (conservative). Attributes 103
Statistical Techniques With the Hurwitz criterion, the decision payoffs are weighted by a coefficient of
optimism (realism), a measure of the decision maker's optimism:

NOTES

The Hurwitz criterion requires that, for each decision alternative, the
maximum payoff be multiplied by the coefficient of optimism, and the minimum
payoff be multiplied by the coefficient of pessimism. In the above example:

For example, assume that the coefficient of optimism is 0.6. Then each of
the weighted values is computed as follows:

The Hurwitz criterion specifies the selection of the decision alternative


corresponding to the maximum weighted value. Therefore, the decision should
be a "large-sized facility" with $206,000.

Association of
104 Attributes
However, if the coefficient of optimism is 0.4, then the values are: Statistical Techniques

NOTES

The decision is now a "Small-sized facility."

The Laplace Criterion


The Laplace criterion is also called the equal likelihood criterion, which
equally weights the possibility of occurrence of each state of nature. For example,
if there are three states of nature, each has a 1/3 chance of occurrence. Then, each
decision alternative will be assigned an average payoff value. In the above
example, we have:

So, the decision is a "Large-sized facility."

Summary of Criteria Results


The decision indicated by the decision criteria examined in the above
example can be summarized as follows:

Association of
Attributes 105
Statistical Techniques The use of several decision criteria often results in a mix of decisions, with
no one decision being selected more than the others. Hence, the appropriate
criterion is dependent on the "risk" personality and philosophy of the
NOTES decision maker.

Decision Making with Probabilities


The Expected Value Criterion
It is possible to obtain probability estimates for each state of nature in
decision-making situations. We use the expected value criterion (used in
Statistics) to identify the best decision alternative. The expected value EV is
calculated by multiplying each decision outcome (payoff value) for each state of
nature by the probability of its occurrence. Let's use the same example as we
used above. This time each of the states of nature is assigned a probability:

We calculate each expected value as


The best decision is the one with the largest expected value. In this case,
$116,000 is the largest value, which gives us the "large-sized facility" as the
best decision.

The Expected Opportunity Loss Criterion


An alternative to the above approach is the expected opportunity loss
criterion (EOL). This utilizes regrets (opportunity losses) to minimize
the expected regret. From the regret table with each state of nature
assigned probability:

We calculate the expected opportunity loss (EOL) for each decision


Association of alternative:
106 Attributes
Statistical Techniques

NOTES

The best decision results from minimizing the regret. In this case, the
decision is a "large-sized facility." The expected value and expected opportunity
loss criteria result in the same decision. You may wonder why you need two
separate approaches to reach the same conclusion. This will be discussed in the
next section.

The Expected Value of Perfect Information


If we examine the best two EV's from the above example,
EV(d1) = $116,000,
EV(d2) = $111,600,
The difference is only $4,400. But if we know there is a poor market
condition, a choice of d2 would be a better choice than d1. Then, how can we
figure out if such a condition may occur? One way is to hire a consultant to do
some research on future market conditions. But it would be foolish for us to pay
a consultant fee more than what we could gain in extra profit from the consultant's
information. The information has some maximum value that represents the limit
of what we would be willing to spend. This value of information can be computed
as an expected value, referred to as the expected value of perfect information
(EVPI). "Perfect information" leads a decision maker to no regret (the right
decision) under each state of nature.

Using perfect information, the expected value of the decision is


$550,000(0.30) + 129,000(0.40) + 0(0.30)
Association of
= $216,600. Attributes 107
Statistical Techniques This is NOT the expected value of perfect information. The expected value
of perfect information is the maximum amount that would be paid to gain
information that would result in a decision better that the one made without
perfect information. The expected value decision without perfect information is
NOTES
to construct a large-sized facility
EV(d1) = $116,000.
The expected value of perfect information is calculated as follows:
EVPI = $216,600 - 116,000 = 100,600.
That is, EVPI is equal to the expected value given perfect information minus
the expected value without perfect information. In this case, $100,600 is the
maximum amount that we would pay to purchase perfect information from other
sources like a consultant and marketing surveys. It is rare for us to obtain perfect
information in general.
It is noted that the expected value of perfect information is the same as the
expected opportunity loss for the decision alternative selected. That is,
EVPI = best EOL.
This is always the case because if we use perfect information, there is no regret.

SUMMARY
In this chapter, we have studied the method of testing association of
attributes between attributes. We have studied the Yule’s association of attributes
method. In this chapter, we have demonstrated the concepts and basics of decision
making with or without probabilities. We have used both a payoff table approach
and a decision tree approach.

REFERENCES
1. https://xisspm.files.wordpress.com/2010/11/decision-theory.doc
2. Statistical and Quantitative methods. M.G. Dhaygude, Himalaya
Publications.

EXERCISE
Q1: Calculate Coefficient of association for the following data.
(i) (AB) = 10, (A) = 24, (αβ) = 8, and N = 40
(ii) (AB) = 12, (A) = 21, (αβ) = 10, and N = 50
(iii) (AB) = 100, (A) = 57, (αβ) = 30, and N = 200

Q2: 300 candidates appeared for a competitive examination and 55 of them


succeeded, 35 received special coaching and out of them 25 candidates
succeeded. Calculate coefficient of association.

Q3: Given (AB) =13, (Aβ) = 24, (αB) = 18, (αβ) = 25 and N = 80. Calculate
Association of coefficient of association.
108 Attributes
Q4: Given (AB) =3150, (A) = 4500, (B) = 6000, and N = 10000. Calculate Statistical Techniques
coefficient of association.

Q5: Given (AB) =13, (Aβ) = 20, (αB) = 15, (αβ) = 25 and N = 100.
NOTES
Calculate coefficient of association.

Q6: In a sample survey of 3500 students, 1500 liked economics, 1250 liked
statistics and 500 liked both economics and statistics. Find coefficient
of association.

Q7: Given (AB) =8, (A) = 18, (αβ) = 5, and N = 35. Calculate coefficient
of association.

Q8. Given N = 160, (A) = 80, (B) = 90, (AB) = 55 Find ultimate class
frequencies.

Q9. Find Yule’s coefficient of association between eye colour of fathers


and their respective sons from the following data.
Fathers with dark eyes having sons with dark eyes = 60
Father with not dark eyes but sons with dark eyes = 80
Fathers with dark eyes but sons with not dark eyes = 90
Fathers with not dark eyes but sons with not dark eyes = 770

Q10. Find Yule’s coefficient of association between extravagance in father


and son from the following.
Extravagant sons with extravagant fathers = 500
Miser sons with extravagant fathers = 210
Extravagant sons with miser fathers = 190
Miser sons with miser fathers = 1250

*****

Association of
Attributes 109

You might also like