FINAL ASSIGNMENT

BUS 173
SUMMER 2022

Group Member Name ID Which Question Answered

Ahmed Morsalin 2131112630 Part 1, Part 3
Khalid Ahmed Shariff 2132279630 Part 2

SUBMISSION DATE: 24th of August, 2022

SUBMITTED TO: RAHNUMA SANJANA, Lecturer, Department of Management
 Answer to the questions in PART 1

Regression Model From EXCEL

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.95369
R Square 0.90952
Adjusted R Square 0.88485
Standard Error 1.69275
Observations 15

ANOVA
Significance
df SS MS F F
Regression 3 316.84445 105.61482 36.85849 4.9462E-06
Residual 11 31.51955 2.86541
Total 14 348.364

Standard Upper Lower Upper

Coefficients Error t Stat P-value Lower 95% 95% 95.0% 95.0%
Intercept 83.72194 3.71607 22.52973 0.00000 75.54294 91.90095 75.54294 91.90095
ADULT MORTALITY
RATE -0.06968 0.01406 -4.95552 0.00043 -0.10063 -0.03873 -0.10063 -0.03873
INFANT DEATH -0.02810 0.00939 -2.99115 0.01227 -0.04877 -0.00742 -0.04877 -0.00742
BMI 0.20520 0.21030 0.97574 0.35016 -0.25766 0.66805 -0.25766 0.66805

a) Life Expectancy = 83.72194 - 0.06968 Adult Mortality – 0.02810 Infant Death + 0.20520 BMI

b) Adult Mortality Rate coefficient is -0.06968, this indicates if adult mortality increases by 1
person per 1000 population, then life expectancy will reduce by 0.06968 years, keeping Infant
Death and BMI constant.

Coefficient with Infant Death is -0.02810 which means if Infant Death increases by 1 person per
1000 population, then life expectancy will reduce by 0.02810 years, keeping Adult Mortality
and BMI constant.
Coefficient with BMI is 0.20520 which means if average BMI point of the entire population
increases by 1 then life expectancy will go up by 0.20520 years, keeping Adult Mortality and
Infant Death constant.

Intercept Coefficient is 83.72194 which means if Adult Mortality rate per 1000 population,
Infant Death per 1000 population and BMI point all three of these are 0, then life expectancy
will be 83.72194 years.

c)

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.95369
R Square 0.90952
Adjusted R Square 0.88485
Standard Error 1.69275
Observations 15

Here from the Regression Statistics Table, we can observe the R2 = 0.90952 which means
90.952% variation in Life Expectancy is explained by Adult Mortality rate, Infant Death and BMI.
Rest (100 – 90.952 = 9.048) 9.048% goes into error.

Hence, we can conclude this is a very good model.

The value of Adjusted R2 is 0.88485 or 88.485%, this means after penalizing the model for
adding more independent variable value of R2 decreased from 90.952% to 88.485%. Total
reduction due to adjustment in R2 is 2.467%.
 Answer to the questions in PART 2

Regression Model From EXCEL

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.93872
R Square 0.88119
Adjusted R Square 0.86139
Standard Error 1.85719
Observations 15

ANOVA
Significance
df SS MS F F
Regression 2 306.97417 153.48708 44.49994 0.00000
Residual 12 41.38983 3.44915
Total 14 348.36400

Standard Upper Lower Upper

Coefficients Error t Stat P-value Lower 95% 95% 95.0% 95.0%
Intercept 58.45598 22.57455 2.58946 0.02368 9.27026 107.64171 9.27026 107.64171
POL3 0.28851 0.22096 1.30571 0.21613 -0.19292 0.76994 -0.19292 0.76994
MEASLES -0.02160 0.01072 -2.01455 0.06692 -0.04495 0.00176 -0.04495 0.00176

d) Life Expectancy = 58.45598 + 0.28851 POL3 – 0.02160 MEASELS

e) Coefficient with POL3 is 0.28851 which indicates if Polio immunization coverage among 1-
year-olds increases by 1 percent then Life expectancy will go up by 0.28851 years, keeping
reported measles cases per 1000 population constant.

Coefficient with Measles is -0.02160 which means if number of reported measles case per 1000
population increases by 1 then Life expectancy will reduce by 0.02160 years, keeping polio
immunization coverage among 1 year-olds constant.

Intercept coefficient 58.45598 means if POL3 is 0%, reported measles case per 1000 population
is 0 then Life expectancy will be 58.45598.
f)

SUMMARY
OUTPUT

Regression Statistics
Multiple R 0.93872
R Square 0.88119
Adjusted R Square 0.86139
Standard Error 1.85719
Observations 15

The Regression Statistics Table gives us the value of R2 which is 0.88119, this means 88.119%
variation in Life Expectancy is explained by POL3 and Measles. Rest (100 – 88.119 = 11.881)
11.881% goes under error.

We can say this is a good model.

The value of Adjusted R2 is 0.86139 or 86.139%, after penalizing the model for adding more
independent variable the value of R2 decreased from 88.119% to 86.139%. It reduced by 1.98%.
 Answer to the questions in PART 3

g)

Regression Model 1 From EXCEL

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.95369
R Square 0.90952
Adjusted R Square 0.88485
Standard Error 1.69275
Observations 15

Regression Model 2 From EXCEL (New model)

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.93872
R Square 0.88119
Adjusted R Square 0.86139
Standard Error 1.85719
Observations 15

To compare different regression models with different number of independent variables

we look into the value of Adjusted R2, here we can observe that the value of Adjusted R2
in model 1 is 0.88485 which is higher than the value of Adjusted R2 in model 2 which is
0.86139. That means model 1 explains (88.485 – 86.139 = 2.346) 2.346% more variation
in Life expectancy than the new model.
Hence, we can conclude that the older model is better than the new model. Because it
explains higher variation in Life expectancy than new model.
h) GENDER could be added as a dummy variable in this model. Because gender have a great
impact on Life expectancy as different gender have few different biological attributes, different
risk factors of diseases which have significant impact on Life expectancy.

So, adding gender as a dummy variable in this model will be appropriate.

 Regression Model 1 From EXCEL work

INFANT
LIFE EXPECTANCY DEATH BMI ADULT MORTALITY RATE
71.8 123 18.3 129
68.3 284 18.7 181
66.4 290 21.6 161
74.9 110 17.2 138
75 121 14.6 123
68.5 265 22.2 211
74.9 112 23.1 148
83.1 101 21.2 55
69.2 214 19.1 165
76 143 17.5 127
68.7 264 19.6 174
69.1 245 21.7 176
66.6 236 21.8 199
77.7 145 17 78
78.5 156 18.5 61

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.95369
R Square 0.90952
Adjusted R Square 0.88485
Standard Error 1.69275
Observations 15.00000

ANOVA
Significance
df SS MS F F
Regression 3 316.84445 105.61482 36.85849 0.00000
Residual 11 31.51955 2.86541
Total 14 348.36400

Standard Upper Lower

Coefficients Error t Stat P-value Lower 95% 95% 95.0% Upper 95.0%
Intercept 83.72194 3.71607 22.52973 0.00000 75.54294 91.90095 75.54294 91.90095
INFANT DEATH -0.02810 0.00939 -2.99115 0.01227 -0.04877 -0.00742 -0.04877 -0.00742
BMI 0.20520 0.21030 0.97574 0.35016 -0.25766 0.66805 -0.25766 0.66805
ADULT MORTALITY RATE -0.06968 0.01406 -4.95552 0.00043 -0.10063 -0.03873 -0.10063 -0.03873
 Regression Model 2 From EXCEL work
LIFE
EXPECTANCY POL3 MEASLES
71.8 85 340
68.3 76 560
66.4 75 520
74.9 86 367
75 87 390
68.5 79 590
74.9 85 389
83.1 96 134
69.2 73 578
76 88 378
68.7 76 589
69.1 79 514
66.6 74 578
77.7 84 323
78.5 85 345

SUMMARY OUTPUT

Regression Statistics
Multiple R 0.93872
R Square 0.88119
Adjusted R
Square 0.86139
Standard Error 1.85719
Observations 15

ANOVA
Significance
df SS MS F F
Regression 2 306.97417 153.48708 44.49994 0.00000
Residual 12 41.38983 3.44915
Total 14 348.36400

Standard Lower Upper

Coefficients Error t Stat P-value Lower 95% Upper 95% 95.0% 95.0%
Intercept 58.45598 22.57455 2.58946 0.02368 9.27026 107.64171 9.27026 107.64171
POL3 0.28851 0.22096 1.30571 0.21613 -0.19292 0.76994 -0.19292 0.76994
MEASLES -0.02160 0.01072 -2.01455 0.06692 -0.04495 0.00176 -0.04495 0.00176