Problem Set 7 Topic:

Multiple Regression

1. Which one of the following statements is TRUE?

A. The SSE always goes up when a nontrivial explanatory variable is added
B. The value of R2 always goes up when a nontrivial explanatory variable
is added to a regression model
C. If we reject H0: β1 = β2 = 0 using the F-test, then we should conclude
that both slopes are different from zero
D. None of the above

2. When the null hypothesis, H0: 1 = 2 = 3 =0, is rejected, the interpretation

should be:
A. There is no linear relationship between y and any of the three
independent variables
B. There is a linear relationship between y and at least one of the three
independent variables
C. All three independent variables have equal slopes
D. There is a linear relationship between y and all three independent
variables

3. Which of the following statements is TRUE?

A. The adjusted R2 always increases as more independent variables are
added to the model while R2 may or may not increase
B. The adjusted R2 is smaller than R2 when the constant term is negative
C. The adjusted R2 adjusts explanatory power by the degrees of freedom
while R2 does not do a similar adjustment
D. The adjusted R2 always decreases as more independent variables are
added to the model while R2 increases.

The next five questions are based on the following information

A consumer interest group surveyed 100 customers who recently purchased a

digital camera. Each customer reported the number of megapixels of the sensor
in the camera (a measure of the highest possible resolution), the price of the
camera (in dollars) and their personal satisfaction (on a scale from 0 to 100).
4. The fraction of the variability of satisfaction explained by the simple
regression of Satisfaction on Megapixels is
A. 0.077
B. 0.142
C. 0.377
D. 0.649
E. 0.874

5. The fact that the coefficient for Megapixels in the multiple regression is
negative, whereas in the simple regression it is positive
A. Shows it should have been added to the model
 B. Means that R2 will reduce in the multiple regression
C. Is not surprising as Price and Megapixels are correlated
D. Is not surprising as Price and Megapixels are uncorrelated

6. The predicted satisfaction of a customer who bought a digital camera for $300
with a 1.5-megapixel sensor is about (Full MRM)
A. 15.25
B. 19.87
C. 34.85
D. 45.31

7. Consider the satisfaction of two customers who both bought digital cameras
for $500. One has a 4-megapixel sensor and the other has a 6-megapixel
sensor. On average, the customer who purchased the camera with 6-
megapixel sensor would report satisfaction (Full MRM)
A. That is identical to the satisfaction of the other customer
B. That is about 2 units less than the satisfaction of the other customer
C. That is about 4 units less than the satisfaction of the other customer
D. That is about 6 units higher than the satisfaction of the other customer
E. That is about 13 units higher than the satisfaction of the other customer

8. The normal quantile plot of the residuals shows that the errors cannot be
taken to be normally distributed. Which of the following statistical
conclusions would most likely be invalid due to this finding?
A. That the mean of the response was 47.1
B. That the partial slope of Megapixels was -1.92
C. That an approximate 90% prediction interval for a new observation
had width of about +/- 22
D. That the proportion of variability explained by the multiple regression
was 44%

The next three questions are based on the following information

An advertising agency wants to understand the determinants of daily viewership of

an advertisement during the IPL season. The agency collects data of views per day
for 30 days and conducts a multiple regression analysis, resulting in the following
summary:

Source DF Sum of Squares

Model * 72.18
Error 24 84.23
C.Total * *
9. How many explanatory variables were considered in determining the daily
viewership?
A. 4
B. 5
C. 6
D. 7

10. What fraction of the total variation in the daily viewership is explained by the
regression model?
A. 0.675
B. 0.539
C. 0.325
D. 0.461

11. The agency wishes to increase the model fit and uses eight regressors instead,
RMSE then decreases by 12% and SST turns out to be 300. What is the R2 of the
new model if the observations are for 61 days?

A. 0.458
B. 0.529
C. 0.595
D. 0.675

An actuarial firm developed a model to predict how long individuals will live.
After consulting a number of physicians, they collected the age at death (y),
the average number of hours of exercise per week (x1), the cholesterol level
(x2), and the number of points that the individual's blood pressure exceeded
the recommended value (x3). A random sample of 40 individuals was
selected. The computer output of the multiple regression models is shown
below.

The Regression Equation is ŷ= 55.8 + 1.79x1 − 0.0168x2 − 0.061x3

Term Std. Error

Intercept 11.8
x1 0.44
x2 0.010
x3 0.014

12. What is the SMALLEST significance level at which we can conclude that
there is a sufficient evidence to infer that the individual's cholesterol level and
the age at death are negatively linearly related fixing other explanatory
variables?
 A. 2.5%
B. 5%
C. 10%
D. 0.1%

(Q13-Q14). Consider the following factors in predicting the top speed of cars for a
sample of 74 observations:
• TOPSPEED: Top speed of the car (in miles per hour)
• WEIGHT: Vehicle weight (in pounds)
• HORSEPOWER: Maximum horsepower of the car’s engine

Summaries of two regressions predicting TOPSPEED are below:

Model 1: Predicted TOPSPEED = 85.24 + 0.00839 WEIGHT

R2 = .313; R2 Adjusted = .304; Residual Standard Deviation = 8.72

Model 2: Predicted TOPSPEED = 96.44 - 0.00792 WEIGHT + 0.349 HORSEPOWER

R2 = .987; R2 Adjusted = .986; Residual Standard Deviation = 1.22

13. Predict the TOPSPEED of a car that weighs 4000 pounds and has 200 horsepower
engine.
A. 64.8
B. 125.4
C. 134.6
D. 192.3

14. Which option below best describes the relationship between WEIGHT and
TOPSPEED?

A. Overall, lighter cars tend to have higher top speeds. For fixed horsepower,
lighter cars still tend to have higher top speeds.
B. Overall, heavier cars tend to have higher top speeds. However, for fixed
horsepower, lighter cars tend to have higher top speeds.
C. Overall, lighter cars tend to have higher top speeds. However, for fixed
horsepower, heavier cars tend to have higher top speeds.
D. Overall, heavier cars tend to have higher top speeds. For a given horsepower,
heavier cars still tend to have higher top speeds.