Download as xlsx, pdf, or txt
Download as xlsx, pdf, or txt
You are on page 1of 20

Top

Sales Price Mileage


Cars speed
(in 1,000 units) (in lakh rupees) (Km/ltr)
Rocinante 1 171.877 6.1 15.8 (Km/hr)
168.2
Rocinante 2 139.796 6.1 12.1 149.6
Rocinante 3 178.947 9.9 17 173.4
Rocinante 4 140.022 5.8 11.6 170.6
Rocinante 5 186.476 10 17.2 175
Rocinante 6 192.123 6.5 17.6 173.1
Rocinante 7 175.085 5.5 16 184.6
Rocinante 8 146.882 8.4 13 175.7
Rocinante 9 202.847 6.6 19.3 166.7
Rocinante 10 149.933 8.8 13.3 175.4
Rocinante 11 171.579 9.5 15.8 150.9
Rocinante 12 201.512 9.2 19.2 140.5
Rocinante 13 142.604 8.1 11.7 150.2
Rocinante 14 135.158 5.4 11.2 146.2
Rocinante 15 107.322 5.6 7.8 151.2
Rocinante 16 199.699 6.3 19 143.2
Rocinante 17 202.875 5.8 19.7 178.9
Rocinante 18 204.406 7.7 19.5 168.2
Rocinante 19 107.331 7.2 7.5 164.7
Rocinante 20 190.345 8.4 18.1 182.7
Rocinante 21 108.419 8.4 7.9 165.7
Rocinante 22 150.249 5.7 12.6 137
Rocinante 23 155.882 5.5 12.9 173.9
Rocinante 24 100.98 8.4 7.6 165.1
Rocinante 25 182.679 5.6 16.7 160.1
Rocinante 26 166.752 9.9 15.5 155.6
Rocinante 27 121.561 6.2 9.5 155.4
Rocinante 28 174.256 9.9 16.3 161.9
Rocinante 29 119.018 6.5 9.1 146.8
Rocinante 30 169.842 7.1 15.3 149.5
Rocinante 31 198.311 9.6 19.1 146.5
Rocinante 32 204.875 5.8 19.4 178.9
Rocinante 33 119.561 6.2 9.4 175.4
Rocinante 34 203.875 5.9 19.4 175.9
Rocinante 35 118.561 6.1 9.4 165.4
Regression Equation : Sales = Constant + b1*Pr
Assumptions : The relationship between depndent variable and of those of the pred

Specifications Rocinante36
Price ( in lakh ) 7
Manufacturing cost ( in lak 6
Mileage ( in km/litre ) 22
Top speed (in km/hr ) 140

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.99788065450456
R Square 0.995765800634449
Adjusted R Square 0.995356039405525
Standard Error 2.30779153196702
Observations 35

ANOVA
df SS
Regression 3 38827.6180187373
Residual 31 165.102954405579
Total 34 38992.7209731429

Coefficients Standard Error


Intercept
Price 50.7231271792463 5.29210288470814
(in lakh rupees)
Mileage -0.795026440875324 0.254333076043226
(Km/ltr) 8.30633109237939 0.100181306818659
Top speed (Km/hr) -0.0185725639929489 0.0305582354491913

Observations
The R-square and the adjusted R Square valures are very high which shows that the fitness of the regr
For the p-value less than 0.05 the variables will be significant and impactful else there wont be any im
the Sales.

Sales = Inercept + b1*Price + b2*M

Sales = 50.7231271792463 -0.795026440875324


The Overall Test The null Hypothesis H0 would be: None of the independent vari
on the dependent varibales.
Hypothesis Testing The Alternate Hypothese Ha is: At least one independent variab
The p-values of theinfluence on the
independent dependent
variables like variable.
price and milage are less than 0.05
The null Hypothesis H0 would be: The independent variable( X1
the dependent variable.
The Alternate Hypothesis Ha would be that : The independent v
The individual Significance Test dependent variab

Intercept
Price
(in lakh rupees)
Mileage
(Km/ltr)
Top speed (Km/hr)
On comparing each of the p-values for the independent variables individually with the cri

Calculations
Equation : Sales = 50.7231271792463 -0.795026440875324

Sales ( in 1000 units )


Profit = Price - Manufacturing cost
( in Lakhs)

Overall Projected Profit = sales * profit


( in crores )

Suppose increasing the price of Rocinante36 and Marengo32 by ₹1 lakh each. Then the bel

Profit =Sales
Price( -inManufacturing
1000 units ) cost
( in Lakhs)
[ sales * profit ]
Profit = Price - Manufacturing cost
New Overall Projected Profit = sales * profit
( in crores )

Droping the insignificant regression variables ( Top speed ) and rebuilding the regression model u

SUMMARY OUTPUT ( After Dropping insignificant Variables )

Regression Statistics
Multiple R 0.997855373536924
R Square 0.995715346496514
Adjusted R Square 0.995447555652547
Standard Error 2.28493913026155
Observations 35

ANOVA
df SS
Regression 2 38825.6506746148
Residual 32 167.070298528013
Total 34 38992.7209731429

Coefficients Standard Error


Intercept
Price 47.7799640191966 2.11344142050695
(in lakh rupees)
Mileage -0.782925327667118 0.251041783157933
(Km/ltr) 8.29446649742739 0.0972879310727833

OBSERVATIONS :
adjusted R-squared value for old regression model
adjusted R-squared value for new regression model

There is a slight increase by 0.0001 in the Adjusted R square val


dropping of the insignificant variables.
Regression Equation : Sales = Constant + b1*Price + b2*Mileage + b3*Top speed
between depndent variable and of those of the predictor vaariables is linear. Observations are independent of each other.
Confidence Level = 95% ( Significance Level is 0.05 )

MS F Significance F
12942.5393395791 2430.11229551562 7.4472712258E-37 => The value is less than 0.05, hence it
5.32590175501867

t Stat P-value Lower 95% Upper 95% Lower 95.0%


9.58468274035523 8.73902387404114E-11 39.929812186171 61.516442172 39.929812186
-3.12592625876236 0.00383346297701066 -1.3137421693289 -0.276310712 -1.313742169
82.9129840301932 5.78356375681691E-38 8.1020099700452 8.5106522147 8.10200997
-0.607776061671793 0.547763026123756 -0.0808964960897 0.0437513681 -0.080896496

re very high which shows that the fitness of the regression line is really good with the model.
e significant and impactful else there wont be any impact, considering that the Intercept, Price and Milage are the only variables having

Sales = Inercept + b1*Price + b2*Mileage + b3*Top speed

Sales = 50.7231271792463 -0.795026440875324* Price + 8.30633109237939 *Mileage


pothesis H0 would be: None of the independent variables has any significant influence
endent varibales.
te Hypothese Ha is: At least one independent variable has a significant
n the
ent dependent
variables like variable.
price and milage are less than 0.05 which means that we can reject the Null hypothesis at 95% confidence level.
pothesis H0 would be: The independent variable( X1, X2, X3 ) doesn’t have a significant influence on
ent variable.
te Hypothesis Ha would be that : The independent variable ( X1, X2, X3 ) has a significant influence on the
variab

P-value Null Hypothesis Significane of the Varibale


8.73902387404114E-11 Rejected Significant
0.00383346297701066 Rejected Significant
5.78356375681691E-38 Rejected Significant
0.547763026123756 Fail to Reject Not Significant
or the independent variables individually with the critical value ( 0.05 ) it can be said that the intercept, Price and Milage are significan

Sales = 50.7231271792463 -0.795026440875324* Price + 8.30633109237939 *Mileage

227.8972261 ( using the price as 7 and mileage as 22 )

2278.972261

nte36 and Marengo32 by ₹1 lakh each. Then the below changes can be observed.

227.8972261

4557.944523

( Top speed ) and rebuilding the regression model using only significant variables.

MS F Significance F
19412.8253373074 3718.25762129514 1.2901829731E-38
5.2209468290004

t Stat P-value Lower 95% Upper 95% Lower 95.0%


22.6076595052896 3.0381260506021E-21 43.475024720316 52.084903318 43.47502472
-3.1187052522431 0.00382734518465502 -1.2942807063832 -0.271569949 -1.294280706
85.2568906128974 2.60256970559406E-39 8.096297466709 8.4926355281 8.0962974667

R-squared value for old regression model 0.9953560394


R-squared value for new regression model 0.9954475557

light increase by 0.0001 in the Adjusted R square value in the new regression model this can be due to the
the insignificant variables.
each other.

less than 0.05, hence it is significant and the regression is possible

Upper 95.0%
61.5164421723217
-0.276310712421805
8.51065221471361
0.0437513681038242

the only variables having an impact on

% confidence level.
nd Milage are significant.
Upper 95.0%
52.0849033180769
-0.271569948951044
8.49263552814579
Top
Sales Price Mileage
Cars speed
(in 1,000 units) (in lakh rupees) (Km/ltr)
(Km/hr)

Marengo 1 20.896 42.5 199.4 9.3


Marengo 2 31.048 36 235.2 9.7
Marengo 3 29.904 54.7 240.8 16.6
Marengo 4 28.792 42.7 232.5 11.7
Marengo 5 16.776 44.9 188.8 13.7
Marengo 6 18.928 35.5 184.2 9.6
Marengo 7 22.776 51.3 207.7 13.7
Marengo 8 36.824 30.4 249.5 12.6
Marengo 9 22.216 38.4 175.8 16.2
Marengo 10 35.456 32.2 245.6 9.6
Marengo 11 28.576 59.4 223.3 11.3
Marengo 12 33.648 50.8 240 13.6
Marengo 13 33.44 57.5 250 13.5
Marengo 14 34.912 44.6 247.4 11.1
Marengo 15 24.016 34.6 192 14.1
Marengo 16 20.992 56.4 201.2 13.4
Marengo 17 21.696 41.4 193.3 12.8
Marengo 18 17.832 53.7 202 16
Marengo 19 25.624 44.8 208.9 12.7
Marengo 20 21.92 32.4 176.9 13.6
Marengo 21 24.792 40.9 218.2 11.6
Marengo 22 22.256 44.1 190 11.4
Marengo 23 15.488 46.6 176.8 16.3
Marengo 24 21.64 31.4 178.9 14.5
Marengo 25 23.264 54.7 192.3 14.3
Marengo 26 19.328 37.9 184.1 9.4
Marengo 27 27.216 52.9 240.3 15.3
Marengo 28 23.096 42.7 192.1 15
Marengo 29 17.824 54.9 199 13.5
Marengo 30 19.144 44 181.1 11.8
Marengo 31 21.6 42.2 181.2 9.3
Regre
The dependent Variable is Sales (in 1000 Units ) where as the indepen
Regression Equation : Sales = Consta
Assumptions : The relationship between depndent variable and of those
Confidence Level = 95

Specifications Marengo32
Price ( in lakh ) 41
Manufacturing cost ( in lakh) 33
Mileage ( in km/litre ) 15
Top speed (in km/hr ) 210

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.929025133175836
R Square 0.863087698072379
Adjusted R Square 0.847875220080421
Standard Error 2.30673039512172
Observations 31

ANOVA
df SS
Regression 3 905.66981774495
Residual 27 143.66713812602
Total 30 1049.336955871

Coefficients Standard Error


Intercept
Price -13.4476488883451 4.4537940289331
(in lakh rupees) -0.186728172116573 0.0554383632622
Top speed (Km/hr)
Mileage 0.22080168203862 0.0172231500844
(Km/ltr) 0.0413011869886482 0.2066403011033

Observations
The R-square and the adjusted R Square valures are high which shows that the
For the p-value less than 0.05 the variables will be significant and impactful else

Sales

Sales = -13.447648888

Hypothesis Testing
The Overall Test The null Hypothesis H0 would be: None of the inde
on the dependent varibales.
The Alternate Hypothese Ha is: At least one indepe
influence on the
The p-values ofdependent variable.
the independent variables like pric
The individual Significance
Test
The null Hypothesis H0 would be: The independen
the dependent variable.
The Alternate Hypothesis Ha would be that : The in
dependent variab
Intercept
Price
(in lakh rupees)
Top speed (Km/hr)
Mileage
(Km/ltr)
On comparing each of the p-values for the independent v

Calculations
Equation :

Sales ( in 1000 units )


Profit = Price - Manufacturing cost
( in Lakhs)
Overall Projected Profit = sales * profit
( in crores )

Suppose increasing the price of Rocinante36 and M

Sales ( in 1000 units )


Profit = Price - Manufacturing cost
( in Lakhs)

New Overall Projected Profit = sales * profit


( in crores )

Droping the insignificant regression variables

SUMMARY OUTPUT ( After Dropping insignificant Variables )

Regression Statistics
Multiple R 0.9289161042461
R Square 0.8628851287278
Adjusted R Square 0.8530912093512
Standard Error 2.2668393794747
Observations 31

ANOVA
df
Regression 2
Residual 28
Total 30

Coefficients
Intercept
Price -12.97013136159
(in lakh rupees) -0.182473820185
Top speed (Km/hr) 0.22013989044

OBSERVATIONS :
Regression Analysis
Sales (in 1000 Units ) where as the independent variables are Price(in lakh rupees) , Mileage (Km/ltr), Top speed (Km/hr)
Regression Equation : Sales = Constant + b1*Price + b2*Mileage + b3*Top speed
p between depndent variable and of those of the predictor vaariables is linear. Observations are independent of each other.
Confidence Level = 95% ( Significance Level is 0.05 )

MS F Significance F
301.889939248317 56.7355100548808 8.74488158910387E-12
5.32100511577839

t Stat P-value Lower 95%


-3.01936928402737 0.00547896316725025 -22.5860793910278
-3.36821221134266 0.00228837673380118 -0.300478297641662
12.820052136582 5.38827494361743E-13 0.185462697105459
0.199869951641267 0.843080033902266 -0.382689688749724

uare valures are high which shows that the fitness of the regression line is really good with the model.
ariables will be significant and impactful else there wont be any impact, considering that the Intercept, Price and Top Speed are the on

Sales = Inercept + b1*Price + b2*Mileage + b3*Top speed

Sales = -13.4476488883451 -0.186728172116573* Price + 0.22080168203862 *Top Speed


l Hypothesis H0 would be: None of the independent variables has any significant influence
dependent varibales.
ernate Hypothese Ha is: At least one independent variable has a significant
ce on the
-values ofdependent variable.
the independent variables like price and topspeed are less than 0.05 which means that we can reject the Null hypothesis at 9

l Hypothesis H0 would be: The independent variable( X1, X2, X3 ) doesn’t have a significant influence on
endent variable.
ernate Hypothesis Ha would be that : The independent variable ( X1, X2, X3 ) has a significant influence on the
ent variab P-value Null Hypothesis Significane of the Varibale
0.00547896316725025 Rejected Significant
0.00228837673380118 Rejected Significant
5.38827494361743E-13 Rejected Significant
0.843080033902266 Fail to Reject Not Significant
g each of the p-values for the independent variables individually with the critical value ( 0.05 ) it can be said that the intercept, Price an

Sales = -13.4476488883451 -0.186728172116573* Price + 0.22080168203862 *Top Speed

n 1000 units ) 25.26484928 ( using the price as 7 and mileage as 22 )


Price - Manufacturing cost
hs) 8
Projected Profit = sales * profit
es ) 2021.187943

se increasing the price of Rocinante36 and Marengo32 by ₹1 lakh each. Then the below changes can be observed.

n 1000 units ) 25.26484928


Price - Manufacturing cost
hs) 9

verall Projected Profit = sales * profit


es ) 2273.836435

oping the insignificant regression variables ( Mileage ) and rebuilding the regression model using only significant variables.

ng insignificant Variables )

SS MS F
905.45725424552 452.72862712276 88.1041690817304
143.879701625447 5.13856077233738
1049.33695587097

Standard Error t Stat P-value


3.69375976376271 -3.51136299897798 0.0015305489196346
0.050303979958483 -3.62742312507697 0.00113000293500281
0.0166095874144 13.2537843925712 1.37964990241042E-13

adjusted R-squared value for old regression model 0.8478752201


adjusted R-squared value for new regression model 0.8530912094

There is a increase by 0.01 in the Adjusted R square value in the new regression model this can be due to the dropping of
variables.
r), Top speed (Km/hr)

dependent of each other.

=> The value is less than 0.05, hence it is significant and the regression is possible

Upper 95% Lower 95.0% Upper 95.0%


-4.3092183856625 -22.58607939 -4.3092183856625
-0.0729780465914833 -0.300478298 -0.0729780465914833
0.256140666971782 0.1854626971 0.256140666971782
0.465292062727021 -0.382689689 0.465292062727021

del.
pt, Price and Top Speed are the only variables having an impact on the Sales.

*Top Speed
can reject the Null hypothesis at 95% confidence level.

e on

ce on the

be said that the intercept, Price and Top Speed are significant.

22080168203862 *Top Speed

n be observed.

nly significant variables.

Significance F
8.30202130911125E-13
Lower 95% Upper 95% Lower 95.0% Upper 95.0%
-20.5364552417575 -5.403807481 -20.5364552417575 -5.40381
-0.285516851992491 -0.079430788 -0.285516851992491 -0.07943
0.186116692958093 0.2541630879 0.186116692958093 0.254163

his can be due to the dropping of the insignificant

You might also like