IMT- PG Programme in Management

Decision Sciences

Random Motors Project

Presented by: Dr. Anuja

Shukla

Agenda
Topic Time (mins.)
Introduction to Random Motors Project- Case Briefing 10
Expectations and Question discussion (Q:1,2,3) 10
Practice: Hypothesis 20
Expectations and Question discussion (Q:4,5,6,7,8) 5
Practice: Regression 20
Common mistakes to avoid 5
Quiz :https://forms.gle/geXfYnWkvGdANYNb6 5
Q&A 15
Total 90

https://www.caradvice.com.au/955136/mitsubishi-loses-supreme-court-appeal-on-misleading-and
deceptive-fuel-consumption-claim/
Introduction to Random Motors Project- Case Briefing
• You are the Lead Business Analyst at Random Motors, an automobile conglomerate that manufactures
cars across a large range of price points
• For the price-conscious working-class consumers, Random Motors manufactures cars under the brand
Rocinante. The various models of Rocinante fall in the range of ₹5–10 lakh. The brand is known for
cars that are value for money and very attractive mileage.
• For the consumers willing to purchase a luxury car, Random Motors operates under the brand
Marengo. Marengo models fall in the range of ₹30–60 lakh. They are considered sports cars and have
high horsepower and top speed.
• Every time Random Motors launches a new model, it advertises the model’s specifications, such as
price, mileage, and top speed, because consumers usually take these factors into account before
purchasing cars.
• Suppose your chief engineer has recently come up with two new models: Rocinante36 and
Marengo32. He promises that they will outperform the existing competitors in the market.
Introduction to Random Motors Project- Case Briefing
• The chief engineer has built 20 prototypes of each of the two models to check if the cars are performing
as per the desired specifications.
• The test drive results for the 20 prototypes of each of the two models are provided in the file titled Data
Exhibit 1. During the test, the mileage and the top speed is recorded for each of the prototypes. The
data will be used for performing hypothesis testing for the claims made on the specifications.
• Currently, the company has enough cash reserves to invest in only one model. Therefore, as the Lead
Business Analyst, you need to carefully choose the right model and decide the number of cars that you
want to produce. Since you believe in making data-based decisions, you ask your team to provide you
with the sales data of all the past models released under the brands Rocinante and Marengo. The sales
data is provided in the file titled Data Exhibit 2.

Specifications Rocinante36 Marengo32

Price ₹7 lakh ₹41 lakh
Manufacturing cost ₹6 lakh ₹33 lakh
Mileage 22 km/litre 15 km/litre
Top speed 140 km/hr 210 km/hr
 Submission guidelines

You need to upload the three files (2 Excel files and One Presentation saved as PDF) in a single ZIP
file. The ZIP file should be named as "Your Name_RandomMotors“
Item Submission Assessment Reference
Data Exhibit 1 Hypothesis_Testing_(Your Name) Q1,2,3
Data Exhibit 2 Regression_Analysis_(Your Name) Q4,5,6,7,8
PPT Your Name_RandomMotors (pdf) Presentation of findings for evaluation

Expectations and Question discussion

✓Q1. 50 (25,25)
 ✓Q2. 100 (50 –p values, 50 right conclusion)
✓Q3. 150 (50- identifying, 100- reason)
✓Q4. 300(50-reg coeff, 50- reg eq, 50-sales)
✓Q5. 150 (75,75)
✓Q6. 50 (identify model)
✓Q7. 100 (50-identify, 50- logic)
✓Q8. 100 (50- Adjusted r value calculation, 50 – Reason for
change) ✓Total =1000 marks
Q-1a) Formulate the null hypotheses to check whether the new models are
performing as per the desired design specifications. 50 Marks (25+25)

For Rocinante36: Mileage HO :

 Top speed HO :
For Marengo32:

Mileage HO :

Top speed HO :

null hypotheses
Q-1b) Formulate the alternate hypotheses to check whether the new models are
performing as per the desired design specifications.
Top speed H1:
For Rocinante36:

Mileage H1:

Top speed H1:

For Marengo32:

Mileage H1: Alternate hypotheses

Q-2) In order to comment on whether the design specifications are being matched or not,
perform relevant hypothesis tests and calculate the p-value for each. What will you
conclude? Assume you are performing the tests at 95% confidence level. 100 (50 –p
values, 50 right conclusion)
For Rocinante36:
 p-value for mileage = p-value for mileage =
p-value for top speed = p-value for top speed =
Conclusion
For Marengo32:
Q-3) You have learnt about the possible errors that might result from the hypothesis tests.
What type of error is more expensive for Random motors based on the hypothesis they
are testing? Why? Assume that you need to refund all your customers if your cars deviate
from specifications. 150 (50- identifying, 100- reason)
Reason:
The type of error which is
more expensive:
 Why?
Type 1 error or Type 2 error
Q-4) Develop a regression equation for each model at 95 percent confidence level. From
the regression equation predict the sales of the two models.
300(50-reg coefficient, 50- regression equation, 50-sales value)
Develop the regression equation for the
Develop the regression equation for the Marengo models and Predict the number of
Rocinante models and Predict the number unit sales of Marengo32 model?
of unit sales of Rocinante36 model?
Regression coefficients
Regression coefficients

Price:
Mileage:

Top speed:

Equation:

Predicted Sales(in units):

Regression
coefficients

Regression equation Sales

Price:

Mileage:
 Top speed:

Equation:

Predicted Sales(in units):

Q-5) Based on sales prediction, what is the overall predicted profit for Rocinante36 model
and Marengo32 model ? 150 (75,75)

Overall predicted profit

Rocinante36 Model:
Profit

Marengo32 Model:
Q-6) As a CEO, you wish to invest only in the model which is predicted to be more
profitable. Which model among Rocinante36 and Marengo32 will you invest in?
(50-identify model)

Which model you will invest in?

Q-7) Now you must have derived the regression equation for both models, Rocinante and
Marengo. Now if you increase the price of Rocinante36 and Marengo32 by 1 lac rupees
each, which car will have a higher impact on the sales due to increase in price? Give proper
logic for your answer. You can consider that all other specifications such as mileage and top
speed remain the same for both models. 100 (50-identify, 50- logic)

Which car is most affected by a price increase? Why?

Q-8) After developing the regression equation for both models (Rocinante and Marengo), if you
analyse the p values for coefficients in the regression results, you will notice that some of the
regression variables (top speed, mileage and price) are insignificant. Remove the insignificant
regression variables from your selection and rebuild the regression model using only significant
variables. Compare the Adjusted R square value for the new and old regression model. Do you
notice any change in Adjusted R square value? If yes, explain the reason for the change. 100
(50- Adjusted r value calculation, 50 – Reason for change).

Is there a change on Adjusted R square Value? If so, Why?

Rocinante Marengo

Old Adjusted Old Adjusted

New Adjusted New Adjusted
R square R square
R square R square

Hypothesis testing

Steps
1. Frame null and alternate hypothesis (one tail vs two
tail) 2. Decide test to be used (z/t )
3. Decide level of significance(or Confidence interval)
4. Run the test to get value of test statistics (Exhibit
1)
5. Calculate p value
(http://courses.atlas.illinois.edu/spring2016/STAT/STAT200/pt.html
) 6. Give result based on p value (reject or fail to reject null) 7.
Frame type of errors, decide which is more costly
Calculation
Method 1 Method 2
Roccinate Mileage p-value http://courses.atlas.illinois.edu/sprin
g2 016/STAT/STAT200/pt.html)
Sample size
1. Use data analysis tool pak
Sample standard de
= 2. Create dummy variable with value 0 3. Use test for
unequal variances to run test at 95% confidence level
Hypothesized mean
Sample mean =average
Std error of mean =standard dev/sqrt of
sample size Rule for decision making

Degrees of freedom N-1 if p≤.05, Reject ho

t-test statistic =(sample mean-hypo if p>.05, Fail to reject ho
mean)/ std error

Calculation
Steps for regression

1. Create multiple regression equation using IDV and

Dv 2. Use exhibit 2 to run regression
3. Check overall significance of f test (or p value) - if p<.05,
reject Ho, means at least one beta is significant -So go to
next table
4. Check p values for individual independent variables 5.
Identify variables with p<.05 i.e. significant. 6. Impute
those beta values in equation created in step 1 7. We
have final regression equation
Calculation
▪ Sales (Q4)
▪ Use regression equation created as per previous slide. Input value of price/tops peed/mileage
(which ever is significant) as per claim to get sales value. (remember sales in thousands)

▪ Profit (Q5)
▪ Use Sales calculated above to calculate profit
▪ Profit= Sales*(price-cost)

▪ Profitable car (Q6)

▪ Use question 5 to answer which car is more profitable

▪ Increase price by 1 lakh rupees (Q7)

▪ Calculate new sales using increased price
▪ Subtract old sales from new sales

▪ R square and adjusted r square (Q8)

▪ Remove insignificant variables and re run the regression equation to find out the change in r and
 adjusted r values. Mention values and reason for change.
Points to remember (or mistakes to avoid)

✓Mention the hypothesis clearly using correct sign

✓Calculate p values using correct tail test (two tail/ one tail)
✓Result should be mentioned as reject or fail to reject null hypothesis
✓Clearly mention type of error and the reason why its more costly. ✓Mention
value of regression coefficients at the provided space in question 4.
✓Mention regression equation with beta value and name of variable.
✓Mention sales (x 1000)
✓Calculate profit (mention rupee/lakh/crores)
✓Compare impact of price on sales (use absolute value-don’t convert in percentage)
✓Mention values of R and adjusted R in question number 8 and also explain reason
for change.
✓Recheck PDF before submission, sometimes text disappears while converting
 Quiz link

• https://forms.gle/geXfYnWkvGdANYNb6

Study material link

https://drive.google.com/drive/u/0/folders/16OuT61
D VUnmgjdHXfU9GbjBYV0Pfwhxp

Mitsubishi case- https://www.caradvice.com.au/955136/mitsubishi

loses-supreme-court-appeal-on-misleading-and-deceptive-fuel
consumption-claim/
Doubts?
All the Best!
https://www.youtube.com/watch?v=Z9Gw9dIJGiA&t=86s&ab_channel=upGrad_Gmba

Regression
 • Regression is a statistical method used to determine the
relationship between variables in the form of an equation.
• In regression terminology, the variable that you predict is
called the ‘dependent’ or the ‘response’ variable. It is
usually denoted by ‘Y’.
• The variable that is used to predict this dependent variable
is called the ‘independent’ or ‘explanatory’ variable. It is
usually denoted by ‘X’.
• In simple linear regression, a regression model is in the
form of a straight line: Y=β0+β1X
• Y = a+bX
Summary
• Regression is the determination of a statistical relationship between
two or more variables.
• In simple regression, we have only two variables, one variable
(defined as independent) is the cause of the behaviour of another
one (defined as dependent variable).

EatingWeight
Gain
 Independent Variable Dependent Variable

Fitted Value vs Residual value

Example
 Hypothesis Testing in Simple Linear Regression

• Null Hypothesis
• Null hypothesis: Temperature does not significantly influence air conditioner sales
• Null hypothesis : β1=0 , where β1 is the coefficient of X in the equation Y=β0+β1X

• Alternative Hypothesis
• Alternative hypothesis : Temperature has a significant influence on the sales
Alternative hypothesis: β1≠0, where β1 is the coefficient of X in the equation
Y=β0+β1X
R square
• R-square, or the coefficient of determination, is a metric used to
quantify the goodness of fit for a regression model.
• R-square explains how close the actual data points are to the fitted
regression line. In other words, you can say that it measures the
percentage of variability in the dependent variable that is explained by
the regression line.
• R-square is equal to the explained variation in the dependent variable
divided by Total variation. R-square values lie between 0 and 1.
R square
• R square=0.8859
• Regression model is able to explain
88.59% of the variability of the
dependent variable
• 60-70% R square shows good
model • Its not adequate and
full proof test. As number of
variables increases, the R
square goes up.
• So, check for standard error.
 Smaller the standard error, more
accurate the predictions of model
The standard error of the estimate

• The standard error of the estimate is

a
measure of the accuracy of
predictions.
The regression line is the line that
minimizes the sum of squared
deviations
of prediction (also called the sum of
squares error), and the standard
error of
the estimate is the square root of the
 average squared deviation.
• Standard error of estimate is the average
prediction error of the regression line
when used to predict dependent variable
from the independent variable.
• It is typically the average prediction error
of this model when it is used for
predicting the air conditioner sales from
the temperatures
• The smaller the standard error the more
accurate the predictions are.
Multiple Linear Regression

• When there are two or more than two independent variables, the
analysis concerning relationship is known as multiple correlation and
the equation describing such relationship as the multiple regression
equation.

Multiple Linear Regression: Hypothesis Testing

 ► Individual significance test

► T test
• Overall Significance Test
• F test
F test- test for over all significance
• Check P value in ANOVA
table
• P<.01, Null hypothesis
rejected at 99%
confidence level
• At least one IDV has
significant impact on DV
Individual significance test : t test
• Check individual p value
• P<.01 reject Null hypothesis for Television advertising and News paper
advertising at 99% CL (i.e. accept Alternate and use Beta values) • P>.01, Fail
to reject Null hypothesis for online advertising at 99% CL (Beta values not
important)
Regression Equation
 R square and Adjusted R square

• R Square is a basic matrix which tells you about that how much variance is been
explained by the model.
• What happens in a multivariate linear regression is that if you keep on adding new
variables, the R square value will always increase irrespective of the variable significance.
• What adjusted R square do is calculate R square from only those variables whose addition
in the model which are significant.
• So always while doing a multivariate linear regression we should look at adjusted R
square instead of R square.
Multiple R 0.8996

R-Square 0.8093

Adj R-Square 0.7572

StErr of Est 1.1610