Statistical Foundation For Analytics-Module 1

Statistical Foundation for Analytics
Dr. P.K. Viswanathan

Professor
Great Lakes Institute of Management,
Chennai, India
Module 1
An Overview of Statistics
Dr. P.K.Viswanathan
Introduction
Managers make sound decisions when they use all
relevant information in an effective and meaningful
manner. The principal purpose of statistics is to
provide decision-makers with a set of techniques
for collecting, analyzing, and interpreting data into
actionable recommendations. Statistical methods
are widely used to aid decision-makers in all
functional areas of management. This module
provides the basic ideas and concepts at a general
level.
Dr. P.K.Viswanathan
1) Why Should I Study

Statistics?
Situation 1
A company has to decide whether to introduce a new product
into the market or not. The company will introduce the product
into the market if 30% of the target audience in the relevant
population will accept the product so that the risk of product
failure is minimized. Obviously consumer acceptance is
paramount in making this decision. To know about the
consumer acceptance in a reasonable manner, the company has
done a "test marketing" exercise. In the test market, 30% of
the sample target audience (based on a sample of 150
consumers) indicate their acceptance of the product. Does the
sample result at 95% confidence level suggest that 30% of the
target audience in the population (entire market) will accept the
product?
Dr. P.K.Viswanathan

Statistics? Continues
Situation 2
A bank which has been steadily losing customers in the light
of intense competition wants to investigate the reasons for
the loss of customers on account of perceived service quality
in critical dimensions like response time, reliability, courtesy
of the service staff, and credibility. The bank would like to
conduct a comprehensive survey to measure the perceived
service quality from the customers' angle on these
dimensions with that of competition. This would help the
bank develop and implement effective strategies to woo its
present customers back as well as to attract new customers.
Dr. P.K.Viswanathan

Statistics? Continues
Can you make the right decision in situation 1 and
situation 2 with minimum risk without the help of
statistics? The answer is clearly a "No". Information
based decision making using statistical analysis is
absolutely essential in the present environment
characterized by intense competition, onslaught of new
products and services, globalization, and revolution of
information technology.
Dr. P.K.Viswanathan
2 ) What is Statistics?
By "Statistics" we mean methods specially
adapted to the collection, classification,
analysis, and interpretation of data for
making effective decisions in all functional
areas of management.
Dr. P.K.Viswanathan
AMEX Gained by Statistical Analysis

American Express Company (AMEX), the pioneer in personal
charge cards during the eighties used to systematically collect
customer feedback data from the marketplace on a continuous
basis. AMEX is well known for its caring attitude towards
customers.
The Analysis and Interpretation of the customer data revealed that
the customers wanted the new card to be processed within three
weeks where as AMEX was taking around 5 weeks. AMEX
decided to issue new cards within two weeks. Similarly another
analysis revealed that the customers wanted the stolen/lost cards to
be replaced within two days where as AMEX was taking two or
more weeks to issue replacement cards. AMEX decided to replace
the lost cards within two days. As a result of these two decisions,
AMEX could generate $1.4million additional profit per year.
Dr. P.K.Viswanathan
3) Typical Application Areas

Quality Management
Materials Management
SQC Techniques
Process Capability
Finance
Inventory Control
Incoming Quality Assessment
Marketing
Financial Ratio Analysis

Cash Forecasting
Marketing Research
Demand Projections
Dr. P.K.Viswanathan
4) Types of Statistics
Descriptive Statistics is
concerned with Data
Summarization,
Graphs/Charts, and
Tables
Inferential Statistics is a
method used to talk about
a Population Parameter
from a Sample. It involves
Point Estimation, Interval
Estimation, and
Hypothesis Testing
Dr. P.K.Viswanathan
Descriptive Statistics
Example
50
45
40
35
30
25
20
15
10
5
0
Machine 1
Machine 2
Machine 3
Machine 4
1st 2nd 3rd 4th
Qtr Qtr Qtr Qtr
The Quality Control Department of a

large manufacturing company would
like to compute summary measures
such as the mean production per shift
for a particular item. The department
would also like to get a comparative
picture of performance of the mean
production across the four machines in
the plant by tabulation. Further the
company might like to graph the
comparative performance of the four
machines by a bar chart depicting the
mean production per shift.
Dr. P.K.Viswanathan
10
Inferential Statistics -Example

Suppose you, as a marketing manager would like to identify a
niche market for your product. You know from your experience
that an accurate assessment of the income of a typical family is
crucial. The average income of this type of families in the
population is estimated by you to be Rs. 320000 based on figures
obtained from a sample. In this example, average income based on
sample is a Point Estimate of the population. The average income
that falls with in a statistically formed interval of 320000 plus or
minus 40000 is called an Interval Estimate. Any statement such
as the average income in the population is more than Rs. 300000
per year is a Hypothesis. As a manager, the interval estimate may
be much more important to you than the rest!
Caution: Inferential Statistics assumes that the sampling
methodology is random (i.e. based on probability sampling)!
Dr. P.K.Viswanathan
11
5) Some Key Terms Used in

Statistics
Population is the collection of all
possible observations of a
specified characteristic of interest.
An example is all the students in
the Quantitative Methods course
in an MBA program.
Sample is a subset of the
population. Suppose you want to
select a team of 20 students from
200 students in an MBA program
for participating in a management
quiz. The total students 200 is the
population. 20 students selected
for the quiz is the sample.
Parameter
is the population
characteristic of interest. For
example, you are interested in the
average income of a particular class
of people. The average income of
this entire class of people is called a
parameter.
Statistic is based on a sample to
make
inferences
about
the
population parameter. If you look at
the previous example, the average
income in the population can be
estimated by the average income
based on the sample. This sample
average is called a statistic.
Dr. P.K.Viswanathan
12
6) Types of Data
Qualitative Data are nonnumeric in
nature and can't be measured.
Examples are gender, religion, and
place of birth.
Quantitative Data are numerical in
nature and can be measured.
Examples are balance in your
savings bank account, and number of
members in your family.
Quantitative data can be classified

into discrete type or continuous
type. Discrete type can take only
certain values, and there are
discontinuities between values,
such as the number of rooms in a
hotel, which cannot be in fraction.
Continuous type can take any
value within a specific interval,
such as the production quantity of a
particular type of paper (measured
in kilograms).
Dr. P.K.Viswanathan
13
7) Types of Data Measurements-Picture
Ordinal
Interval
Ratio
Dr. P.K.Viswanathan
Information Content
Increases
Nominal
14
7) Types of Data Measurements

Nominal Data: The weakest data
measurement. Numbers are used to
label an item or characteristic.
Categorization is the main purpose
of this measurement. Examples: A
business school may designate
subject specialization by numbers,
i.e., MBA in Finance =1, MBA in
Systems = 2. Various brands of
toothpaste; savings bank account
numbers are other examples of
nominal data. Note that nominal
data cannot be manipulated in a
numerical fashion.
Ordinal or Rank Data: Numbers

are used to rank. An example is
Customer Preference for your brand.
An average preference is rated at 3,
a strong preference at 5. Simple
arithmetic operations are not
possible for ordinal data. Ordinal
data can also be verbalized on a
continuum like excellent, good, fair
and poor. In ordinal data, distance
between
objects
cannot
be
measured.
Dr. P.K.Viswanathan
15
7) Types of Data Measurements

Continues
Interval Data: If you have data
with ordinal properties and can also
measure the distance between
objects, you have an interval
measurement. Interval data are
superior to ordinal data because,
with them, decision makers can
measure the distances between
objects. For example, frozen-food
distributors are concerned with
temperature, which is an interval
measurement. Interval data have
arbitrary
zero
point.
Basic
arithmetic operations are possible
with interval data
Ratio Data: It is the highest level of

measurement and allows you to
perform all basic arithmetic
operations, including division and
multiplication. Data measured on a
ratio scale have a fixed zero point.
Examples include business data,
such as cost, revenue, market share
and profit.
Dr. P.K.Viswanathan
16
8) Data Sources
Primary Data are collected by the organization itself for a
particular purpose. The benefits of primary data are that they fit
the needs exactly, are up to date, and reliable.
Secondary Data are collected by other organizations or for other
purposes. Any data, which are not collected by the organization
for the specified purpose, are secondary data. These may be
published by other organizations, available from research studies,
published by the government, and so on. Secondary data have the
advantages of being much cheaper and faster to collect. They also
have the benefit of using sources, which are not generally
available. Companies will, for example, respond to a survey by
the Government of India, or Confederation of Indian Industry, but
they would not answer questions from another company.
Dr. P.K.Viswanathan
17
9) A Step-By- Step Approach to

Statistical Investigation
Dr. P.K.Viswanathan
18

Statistical Foundation For Analytics-Module 1

Uploaded by

Copyright:

Available Formats

Statistical Foundation For Analytics-Module 1

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Statistical Foundation For Analytics-Module 1

Uploaded by

Copyright:

Available Formats

Statistical Foundation for Analytics

Dr. P.K. Viswanathan

1) Why Should I Study

1) Why Should I Study

1) Why Should I Study

AMEX Gained by Statistical Analysis

3) Typical Application Areas

Financial Ratio Analysis

The Quality Control Department of a

Inferential Statistics -Example

5) Some Key Terms Used in

Quantitative data can be classified

7) Types of Data Measurements-Picture

7) Types of Data Measurements

Ordinal or Rank Data: Numbers

7) Types of Data Measurements

Ratio Data: It is the highest level of

9) A Step-By- Step Approach to

You might also like