BUDDHA SERIES

(Unit Wise Solved Question & Answers)

Course – MBA
COLLEGE – BUDDHA INSTITUTE OF MANAGEMENT
(AKTU Code-1212)

Department of Business Administration

Subject: Business Statistics & Analytics
Subject Code: BMB 104
Faculty Name: Abhay Kumar Upadhyay
 UNIT 3

1. Explain the concept of correlation.

Answer:

Correlation refers to a process for establishing the relationships between two variables. Methods of
correlation summarize the relationship between two variables in a single number called the correlation
coefficient. The correlation coefficient is usually represented using the symbol r, and it ranges from -1
to +1.

A correlation coefficient quite close to 0, but either positive or negative, implies little or no relationship
between the two variables. A correlation coefficient close to plus 1 means a positive relationship
between the two variables, with increases in one of the variables being associated with increases in the
other variable.

A correlation coefficient close to -1 indicates a negative relationship between two variables, with an
increase in one of the variables being associated with a decrease in the other variable. A correlation
coefficient can be produced for ordinal, interval or ratio level variables, but has little meaning for
variables which are measured on a scale which is no more than nominal.

In statistics, Correlation studies and measures the direction and extent of relationship among variables,
so the correlation measures co-variation, not causation. Therefore, we should never interpret correlation
as implying cause and effect relation. For example, there exists a correlation between two variables X
and Y, which means the value of one variable is found to change in one direction, the value of the other
variable is found to change either in the same direction (i.e. positive change) or in the opposite direction
(i.e. negative change). Furthermore, if the correlation exists, it is linear, i.e. we can represent the relative
movement of the two variables by drawing a straight line on graph paper.

1. Rank Correlation (Spearman’s Rank Correlation)

Formula (when no ties):

( ²)
ρ= 1−
( ² )

When ranks include ties:

1
( 12
(m₁³−m₁ + m₂³−m₂ + .……..)
ρ= 1−
( ² )

2. Karl Pearson’s Coefficient of Correlation

Formula:

r= where as X − X = dx , Y − Ȳ = dy
√ ( ²× ²]

On the basis of standard deviation

Shortcut formula:

× ×
r=
( × (∑ ) )× ( × (∑ ) )

Standard deviation formula

r= × ×

2. Distinguish between regression and correlation.

Answer:

Difference between Correlation and Regression

Parameter Correlation Regression

Correlation measures the degree of Regression is about how one variable
Definition
relationship between two variables. affects the other.
To find the numerical value that defines To estimate the value of a random
Objective and shows the relationship between two variable based on the values of fixed
variables. variables.
It doesn’t capture causality but the It is based on causality and shows no
Causality degree of interrelation between two degree of connection but cause and
variables. effect.
Dependent and Both variables are different (One variable
No Differences (both the variables are
Independent is a random variable while another is a
random variables).
Variables fixed variable).
Output will be the same if variables are Output will be changed if variables are
Interchangeable
interchanged. interchanged.
The Coefficients are generally relative The coefficient is generally an absolute
Coefficient
measures. measure.

3. What is regression? Explain the features of correlation coefficient.

Answer:

Regression analysis refers to assessing the relationship between the outcome variable and one or more
variables. The outcome variable is known as the dependent or response variable and the risk
elements, and co-founders are known as predictors or independent variables. The dependent variable is
shown by “y” and independent variables are shown by “x” in regression analysis.

Some of the properties of a regression coefficient are listed below:

 The regression coefficient is denoted by b.

  The regression coefficient of y on x can be represented as byx. The regression coefficient of x on y
can be represented as bxy. If one of these regression coefficients is greater than 1, then the other
will be less than 1.

 They are not independent of the change of scale. They will change in the regression coefficient if
x and y are multiplied by any constant.

 The arithmetic mean of both regression coefficients is greater than or equal to the coefficient of
correlation.

 The geometric mean between the two regression coefficients is equal to the correlation coefficient.

If bxy is positive, then byx is also positive and vice versa.

4. Explain partial and multiple regression.

Answer:

Partial Correlation: Two variables, A and B, are closely related. The correlation between them is
partialled out, or controlled for the influence of one or more variables is called as partial correlation. So
when it is assumed that some other variable is influencing the correlation between A and B, then the
influence of this variable(s) is partialled out for both A and B. Hence it can be considered as a
correlation between two sets of residuals. Here we discuss a simple case of correlation between A and B
is partialled out for C. This can be represented as AB C. r which is read as correlation between A and B
partialled out for C. the correlation between A and B can be partialled out for more variables as well.

Multiple correlation: The multiple correlation coefficient denoting a correlation of one variable with
multiple other variables. The multiple correlation coefficient is denoted as R ABCD…K which denotes that
A is correlated with B, C, D, up to k variables. For example, we want to compute multiple correlation
between A with B and C then it is expressed as [Link]. In this case we create a linear combination of the
B and C which is correlated with A.

The Multiple R can be calculated for two predictor variable as follows.

5. Explain the application of regression and correlation

Answer:

Applications of Correlation and Regression

Before delving into their practical applications, it's important to understand the significance of correlation
and regression analysis. These statistical methods allow us to quantify relationships between variables
and make predictions based on observed data. Correlation measures the strength and direction of
association between two variables, while regression analysis helps in understanding and predicting the
relationship between a dependent variable and one or more independent variables.

Here are some real-life applications of correlation and regression analysis:

 Economics and Finance

 Healthcare and Medicine

 Social Sciences

 Engineering and Technology

 Environmental Science

Economics and Finance

Correlation and regression analyses are integral tools in economics and finance, enabling investors,
economists, and marketers to make data-driven decisions and predictions to optimize portfolio
performance, economic policy, and marketing strategies.

In Stock Market Analysis

 Portfolio Diversification: Investors rely on such correlation analysis to reveal the correlation
between various stocks or assets held in the portfolio. Through discrimination of assets which have
a low or even negative correlation with each other, a pattern of investment dives into diversified
portfolios which decrease risk cumulatively.

 Risk Management: Regression analysis can be the tool that enables investors to put a number on
how one stock’s or asset’s return is associated with several indicators like market indices, interest
rates, or macro indicators. Such assessments enable investors to analyze these factors as potential
performance levers and adjust their risk exposure if the need arises. 118 Marathi

In Economic Forecasting

 GDP Growth Forecasting: With multiple regression analysis, future GDP growth is forecasted
where historical GDP data and pertinent economic indicators including that of consumer spending,
investment, government expenditure, as well as net exports are used. The economists put them
together and can produce models or equations to help them forecast possible future economic
performances.

 Inflation Rate Forecasting: Furthermore, regression can be useful in predicting the inflation rates
through analysis of the historical data, inflation channels like money supply growth, interest rates,
and labor market conditions. The understanding of this gives policymakers and investors the
 potential to predict coming out such trends in the future hence enabling the formulation of sound
economic policies as a result.

 Unemployment Rate Forecasting: A regression analysis is also employed in predicting

unemployment rates by examining data in relation to unemployment rate and all its determinants,
which include economic growth, shrinking or expanding labor force participation, and demographic
developments. Through specifying what are the forces that lead to unemployment policy makers
may design and erect the directed measures to combat the labor market imperfections.

In Marketing ROI

 Advertising Effectiveness: The purpose of regression analysis is to examine how efficiently

marketing campaigns are conducted. These performance metrics are selected based on advertising
expense and sales revenue, brand awareness, or customer engagement. Through measuring the
adverse effects of advertising that stem from these consequences, marketers can apply a corrective
action to advertising approaches and drive increase in the potential returns on investment (ROI).

 Market Segmentation: Correlation analysis becomes an important tool also to tailor marketing
campaigns through different market segments with different responsive rates. The correlation
marketing can be traced down to determinants of customer demographics, psychographics or buying
behavior and comparison outcomes leading to choice of marketing strategies that suit well specific
segments of customers and performance.

Healthcare and Medicine

In healthcare and medicine, correlation and regression analyses play crucial roles in improving patient
outcomes, optimizing treatments, and informing preventive measures. Here are real-life examples of their
application:

In Clinical Trials

Example: A pharmaceutical company got a permission to compare a new drug or medicinal drops in
lowering blood pressure of hypertensive patients in a clinical trial.

 Application of Regression Analysis: Regression analysis is applied by the investigators in

analyzing the data obtained during a clinical trial. They focus on how the dosage of the drug, its
frequency of administration, and the patients' classifications (for example, gender, age, weight) are
associated with the changes in blood pressure occurring during the time of the study. This allows
doctors to determine the specific effective dosage and appropriate treatment duration and regimen
that will bring the most desired therapeutic effects without the adverse effects, or side effects.

In Disease Risk Assessment

Example: Public health officials set a target among themselves to evaluate all the factors of the risk of
developing cardiovascular diseases (CVDs) in that area.

 Application of Correlation Analysis: The analyzing team relies on correlation analysis, which is
a way to uncover relationship between lifestyle factors (for example: diet, exercise habits, smoking
status) and health outcomes that are related to CVD like incidence rates, mortality rates, and
biomarkers (for example: cholesterol levels, blood pressure). This review points out that certain
behaviors like unhealthy eating habits and physical inactivity are very likely to be linked with CVD
risk factors.
  Application of Regression Analysis: Regression is the next step, with the objective of creating
predictive models that estimate an individual’s risk of getting CVD on the basis of their lifestyle
factors using these modelling techniques. These models calculate the dose of association of the risk
factors and the disease end outcomes thus instead of prescription intervention strategies such as
providing targeted methods that promote healthy behaviors and reduce modifiable risk factors are
made possible.

Social Sciences

Correlation and regression analyses are valuable tools in Social Sciences for identifying patterns,
understanding relationships between variables, and informing decision-making processes.

In Education

Regression Analysis in Academic Achievement: A school district is concerned about the student
achievements but not sure of how to solve the issue. They generate data on different areas of need such
class size, teacher experience, student to teacher ratio and student demographics. Through applying a
regression analysis, they can determine the cutting edge factors behind the increase of academic
performance. Such as they get to discover that a class of fewer students and more experienced teachers
may very well relate with better students’ outcomes. The information could be used to facilitate resource
allocation, and policy makers could utilize this insight to improve learner outcomes.

In Sociology

 Correlation Analysis in Crime Rates: A sociologist tries to explain the causes that may be found
to be useful in understanding the difference in the level of crime across neighborhoods. They
compile data on factors like income bracket, education, and unemployment percentage as well as
police presence. They can conduct correlation studies so as to establish the existing relationships
between gun violence and select social factors. This may be illustrated through crime analysis
information where regions that include low level of income and high level of unemployment have
higher chances of crime. The resulting awareness can help formulate social policies that will direct
towards crime reduction by tackling the core social-economic problems.

 Correlation Analysis in Voting Behavior: In the election campaign, it is analysts of political

scientists who want to get an idea of these factors that influence the voting behavior. They then
gather data on different variables including the endorsement candidates, political parties that are
often involved, socioeconomic level and also media coverage. One tool that the campaigns have at
their disposal is correlation analysis, which allows them to explore the relations between these
variables and the election results. They may discover that socioeconomic status and media coverage
as factors that influenced voting scheme. This insight enables the political strategists to construct
their messages as well customize them for specific demographic audiences of voters with which it
is intended to connect an impact.

Engineering and Technology

Real-life applications of correlation and regression in Engineering and Technology, particularly in the
areas of Quality Control and Predictive Maintenance are:

In Quality Control

Example: For automotive manufacturing, regression analysis has been applied to correlate processing
variables like pressure, temperature, and material composition with product qualities like rigidity,
durability and dimensional precision. Through a data collection process that features information from
the various stages of the production cycle and submitting it for regression analysis, manufacturers conduct
an assessment of the process parameters that they find to be close to being critical and end up significantly
influencing the quality of product.

In Predictive Maintenance

Example: One may face a situation where safety and efficiency are of the utmost importance and
regression analysis is a powerful tool used for predictive maintenance programs applied to aircraft
engines. Using criteria such as motor operating conditions, fuel consumption, and vibration levels, the
airline can collect a huge amount of historical data not just on motor performance but also on a running
engine. With this process of regression analysis, the resulting forecast models will be equipped to assess
potential failures or maintenance requests accurately before they take place.

Environmental Science

Correlation and regression analyses are invaluable tools in environmental science for understanding
complex relationships between environmental variables and ecosystem dynamics. Whether modeling
future climate scenarios or assessing the impact of habitat fragmentation on biodiversity, these statistical
techniques provide essential insights that inform decision-making and conservation efforts aimed at
preserving our planet's ecosystems and mitigating the effects of climate change.

Conclusion

In conclusion, correlation and regression are like detectives in the world of numbers, helping us
understand how things are connected. Correlation gives us a sense of how two things change together,
while regression dives deeper, predicting what might happen next. Whether it's figuring out where the
stock market is headed or how our actions affect the environment, these tools are essential.
 Problem 1 : Calculate 𝐗, Ȳ 𝝈𝒙 𝝈𝒚 and r from the following data

X 58 50 53 60 63 55 60 59 61 51
Y 115 110 121 120 124 112 118 115 118 117

Solution : Table for calculation of r

X Dx= X- X Dx 2 Y Dy= Y- Ȳ Dy2 DxDy
58 1 1 115 -2 4 -2
50 -7 49 110 -7 49 49
53 -4 16 121 4 16 -16
60 3 9 120 3 9 9
63 6 36 124 7 49 42
55 -2 4 112 -5 25 10
60 3 9 118 1 1 3
59 2 4 115 -2 4 -4
61 4 16 118 1 1 4
51 -6 36 117 0 0 0
∑X=570 ∑dx= 0 ∑Dx2=180 ∑Y=1170 ∑Dy= 0 ∑Dy2=158 ∑DxDy = 95

∑X
𝑋= = 570/10 =57
∑Y
𝑌= =1170/10 = 117

∑Dx2 180
𝜎 = = = √18 = 4.24
N 10

∑Dy2 158
𝜎 = = = √15.8 = 3.97
N 10

Karl pearson coefficient of correlation

r=
× ×

=
× . × .

= = 0.56 Ans
.
Problem 2 : Find the Karl pearson’s coefficient of correlation on the basis of data ahead
related to age of husband and their wives taking deviation from actual means 30 and
25 respectively :

Couple Age of Husband (in years) Age of Wives ( in Years)

1 23 18
2 ? 20
3 28 22
4 28 27
5 29 ?
6 30 29
7 31 27
8 ? 29
9 35 28
10 36 29

Solution :

Suppose the age of husband of second couple be a

Hence the age of husband of 8th couple = a+6 years

Now mean of x = 30
∑X 23+a+28+28+29+30+31+a+6+35+36
𝑋= =
2a+246
30 =

30x10 = 2a + 246

2a= 300-246

2a= 54
54
a=

a= 27

Now mean of Y = 25
∑Y 18+20+22+27+b+29+27+29+28+29
𝑌= =
b+229
25 =

25x10 = b + 229
b= 250-229

b = 21

Putting value of a, a+6 , & b in their place we have

Table for calculation of r

X Dx= X- X Dx 2 Y Dy= Y- Ȳ Dy2 Dx Dy
23 -7 49 18 -7 49 49
27 -3 9 20 -5 25 15
28 -2 4 22 -3 9 6
28 -2 4 27 2 4 -4
29 -1 1 21 -4 16 4
30 0 0 29 4 16 0
31 1 1 27 2 4 2
33 3 9 29 4 16 12
35 5 25 28 3 9 15
36 6 36 29 4 16 24
∑X=300 ∑dx= 0 ∑Dx2=138 ∑Y=250 ∑Dy= 0 ∑Dy2=164 ∑DxDy= 123

Now K.P.C of correlation is

r=
√ ( ²× ²]

=
√ ×

=
√

= .

= 0.8176 ≈ 0.82
Problem 3 : The coefficient of correlation between X and Y series is +0.48. Their
covariance is 36. The variance of x is 16. Find the standard deviation of Y series.

Solution : Given r= +0.48 , Covariance = 36 , Vx = 16 , σy = ?

Now by formula

r=
×

Vx= (σx)2
σx= √16

σx= 4

0.48 =
×

σy =
× .

=
.

= 18.75 answer

Problem 4 : For a sample of 12 observation from a data given below find coefficient
of correlation:

∑X=120 ,∑Y=432, ∑XY= 4992, ∑Y2= 18252, ∑X2= 1392

Solution :
× ×
r=
( × (∑ ) )× ( × (∑ ) )

× ( × )
r=
( × ( ) )× ( × ( ) )

r=
( )× ( )

r=
√ ×

=
×

=+0.933 ans

Problem 5: Comment on the coefficient of correlation if N=100 and P.E= 0.024

Solution :

Formula P.E = 0.6745 ×

√

0.024 = 0.6745 ×
√
 1 − 𝑟 = 0.364

r2 = 1-0.36 = 0.64

r= 0.8

Problem 6: Calculate rank correaltion coefficient from the following data

Price of Tea Price of Coffee

75 120
88 130
95 150
70 115
60 110
80 140
81 142
50 100
Solution : Table for calculation of ρ
Price of Tea (X) Rank (Rx) Price of Coffee Rank (Ry) Difference of D2
(Y) rank(Rx- Ry)
75 5 120 5 0 0
88 2 130 4 -2 4
95 1 150 1 0 0
70 6 115 6 0 0
60 7 110 7 0 0
80 4 140 3 1 1
81 3 142 2 1 1
50 8 100 8 0 0
N=8 N=8 ∑ D 2= 6

Rank Correlation (Spearman’s Rank Correlation)

Formula (when no rank repeat):

( ²)
ρ= 1−
( ² )

( × )
=1 −
( ² )

=1−
( )

=1−
×

=1−

= 504 − =1− = 0.928 ≈ 0.93