Homework Assignment Meeting 5

1. A line has been fit to data representing cholesterol readings for 28 individuals starting a
cholesterol-reducing drug. The computer provides the following output.

The degrees of freedom for the t test for the slope are:

a. 28
b. 27
c. 26
d. 25

2. Which of the following statements about simple linear regression is FALSE?

a. The term residual refers to the difference between the observed response and the
predicted response using the regression equation.
b. If b0 and b1 are the intercept and slope, respectively, of the least squares line, then
yˆ = b0 + b1 x *
is an unbiased estimator of the mean response when x = x∗.

s=
 ( y − yˆ )
i i
2

c. The estimate of σ is given by n −1 .

d. The ei sum to zero.
e. The ei are the vertical deviations of the observed points from the least squares line.

3. A researcher measured the Height (in feet) and Volume of usable lumber (in cubic feet) of 25
oak trees. The goal is to determine if the volume of usable lumber can be estimated from the
height of a tree. From the computer output for the least squares fit of Volume on Height, the
following results were given:

r = 0.40

 (x − x ) = 374.46 y = 40 å( y - y ) = 1837.5
2
x = 75
2
i i

What is the equation of the least squares regression line?

a. Volume = 67.78 + 0.181 Height

b. Height = 67.78 + 0.181 Volume
c. Volume = -107.21 + 1.963 Height
d. Volume = -26.46 + 0.886 Height
e. Volume = 10 + 0.4 Height
4. A line has been fit to data representing cholesterol readings for 28 individuals starting a
cholesterol-reducing drug. The computer provides the following output.

The 95% confidence interval for the slope is:

a. 0.6627 ± 2.055 (0.1428).

b. 0.6627 ± 1.96 (0.1428).
c. 0.6627 ± 0.1428.

5. Confidence intervals for a mean response are narrowest where?

a. Near the ends of the data
b. Near the middle of the data
c. They are the same width throughout a data set.

6. A multiple regression was performed for a data set with 4 variables and 84 observations. The
P-value for the F test was p = 0.001. This means that:
a. all the slopes in the model are non-zero.
b. at least some of the slopes in the model are non-zero.
c. none of the slopes in the model are non-zero.

7. A forester wishes to predict the volume (in cubic feet) of usable lumber in a certain species of
tree using the height (in feet) and diameter (in inches) of the trees. The height and diameter
of 31 trees of a certain species were measured, the trees were cut down, and the volume of
usable lumber was determined. The explanatory variable(s) in this study is/are:
a. either height or diameter.
b. both height and diameter.
c. height, diameter, and volume.
8. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

Based on these results, a 90% confidence interval for the coefficient of diameter is:

a. 4.708 ± 1.645 (0.2643).

b. 4.708 ± 1.701 (0.2643).
c. 4.708 ± 2.048 (0.2643).

9. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

The value of the MSregression for this model is:

a. 7684.2.
b. 3842.1.
c. 15.07.
10. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

Based on these results, a 99% confidence interval for b0 is found to be -57.99 ± 23.87. We
may conclude that:

a. if we tested the hypotheses H0: b0 = 0, Ha: b0 ≠ 0, the P-value would be less than 0.01.
b. the estimate of b0 would tell us the volume of a tree of this species when height and
diameter are 0.
c. Both of the above

11. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

The value of the F statistic for this model is:

a. 18.21.
b. 254.95.
c. 14.22.

12. The degrees of freedom for the F test are:

a. 2 and 28.
b. 2 and 30.
c. 28 and 30.
13. What determines the amount of college tuition students at different schools have to pay? A
random sample of American colleges and universities is selected, and the values of several
variables are recorded: Average Annual Tuition (Y), Percent of students in the Top 25% of
their high school graduating class (X1), Student-to-Faculty Ratio (X2), Average Faculty
Compensation (X3), Acceptance Rate (X4), and Percentage of Faculty with a PhD (X5). How
many parameters will be in the multiple least squares regression equation (including the
intercept)?
a. 5
b. 6
c. 7
d. 8

14. What determines the amount of college tuition students at different schools have to pay? A
random sample of American colleges and universities is selected, and the values of several
variables are recorded: Average Annual Tuition (Y), Percent of students in the Top 25% of
their high school graduating class (X1), Student-to-Faculty Ratio (X2), Average Faculty
Compensation (X3), Acceptance Rate (X4) and Percentage of Faculty with a PhD (X5). The
regression equation is as follows:

yˆ = −2081 .2 + 113 .2 x1 − 293 .4 x2 + 0.12 x3 + 6619 .7 x4 + 15 .9 x5

if the % of students that graduated in the top 25% in high school is 35, the student−faculty
ratio at this university is 19.1, faculty compensation is $46,000, there is a 0.821 acceptance
rate, and 89% of faculty have a PhD. What is the predicted average annual tuition?

a. 7204.02
b. 9374.23
c. 8646.73

15. If the observed value for the average annual tuition is at the particular university in the
above slide is $6198.00, what is the residual value for that university?
a. 3447.92
b. -2448.73
c. 2448.73
d. -3447.92
16. What determines the amount of college tuition students at different schools have to pay? A
random sample of American colleges and universities is selected, and the values of several
variables are recorded: Average Annual Tuition (Y), Percent of students in the Top 25% of
their high school graduating class (X1), Student-to-Faculty Ratio (X2), Average Faculty
Compensation (X3), Acceptance Rate (X4) and Percentage of Faculty with a PhD (X5). If we
sampled 18 American universities, how many degrees of freedom associated with error will
be in the ANOVA table?
a. 5
b. 10
c. 12
d. 14

17. On each of 17 days, a researcher studying the effects of firefly’s dependence on external
environmental factors collected the following data. The researcher was interested in the
time of the first flash of the fireflies. The explanatory variables she considered were light
intensity and temperature. In conducting an overall F test, what are the degrees of freedom
for regression?
a. 1
b. 2
c. 3
d. 4

18. what are the degrees of freedom for error?

a. 14
b. 15
c. 16
d. 17

19. In conducting an overall F test at the 5% level of significance, the null hypothesis will be
rejected if F >:
a. 2.73.
b. 3.74.
c. 4.86.
20. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

The value of the multiple R2 is:

a. 0.0549.
b. 0.0520.
c. 0.9480.

21. A multiple regression was performed on data obtained from 31 trees to predict usable
volume from the height (in feet) and the diameter (in inches) of the tree. Partial computer
results are shown below.

Based on these results, the researcher decides to test the hypotheses H0: b2 = 0, Ha: b2 ≠ 0
and finds the p -value for this test to be less than 0.02. We may conclude that:

a. height would be statistically significant at the 0.02 level as an explanatory variable in a

simple linear regression model with volume as the response.
b. since height is statistically significant at the 0.02 level, we probably do not also need
diameter as an explanatory variable.
c. Neither (a) nor (b)

22. All three parameters are found to be statistically significant at the 0.05 level. In addition, the
value of R2 for this multiple regression model is found to be 0.948. We may conclude that:
a. the model fits quite well and we have verified the validity of the multiple linear
regression model.
b. the model fits quite well and all the observed volumes in the data can be predicted with
0.948 accuracy from a model using both height and diameter.
c. Neither of the above
23. A multiple regression was carried out with 64 cases and five variables. SSM was 2400 and SSE
was 130. Recall: SST = SSM + SSE. The squared multiple correlation for the model is:
a. R2 = 0.054.
b. R2 = 0.949.
c. R2 = 0.895.

24. We have data on 143 brown bears and construct a model to predict weight based on head
length and width, neck girth, length of the body, and chest girth. Partial regression output is
below.

According to this information, which variables do not significantly contribute to predicting a

bear’s weight?

a. All of them contribute.

b. Head length and width and body length do not contribute.
c. Neck girth and chest girth do not contribute.
25. We have data on 143 brown bears and construct a model to predict weight based on head
length and width, neck girth, length of the body, and chest girth. A plot of residuals against
fitted values is shown to the right. Does it indicate any problems with the regression model
as fit?

a. There are no problems.

b. Yes, the plot indicates there is nonconstant variance around the line.
c. Yes, the plot indicates curvature is present.
 Answers

1. C
2. C
3. D
4. A
5. B
6. B
7. B
8. B
9. B
10. A
11. B
12. A
13. B
14. C
15. B
16. C
17. B
18. A
19. B
20. C
21. C
22. C
23. B
24. B
25. C