Solution Manual: Homework 3

1. (a) $23,400 (recall that Price is measured in $1000).

(b) In this case ∆BDR = 1 and ∆Hsize = 100. The resulting expected change in
price is 23.4 + 0.156 × 100 = 39.0 thousand dollars or $39,000.
(c) The loss is $48,800.
n−1 n−k−1
(d) Since R̄2 = 1 − n−k−1
(1 − R2 ), so R2 = 1 − n−1
(1 − R̄2 ) = 0.727.

n−1
2. (a) Since R̄2 = 1 − n−k−1 (1 − R2 ), we can calculate the values of R̄2 are 0.162,
0.180 and 0.181 for columns (1) to (3).
(b) i. Workers with college degrees earn $8.31/hour more, on average, than work-
ers with only high school degrees.
ii. Non-female earn $3.85/hour more, on average, than female.
(c) i. On average, a worker earns $0.51/hour more for each year they age.
ii. Sally’s earnings prediction is 1.87 + 8.32 × 1 − 3.81 × 1 + 0.51 × 29 = 21.17
dollars per hour. Betsy’s earnings prediction is 1.87 + 8.32 × 1 − 3.81 ×
1 + 0.51 × 34 = 23.72 dollars per hour. The difference is $2.55/hour.
(d) i. Workers in the Northeast earn $0.18 more per hour than workers in the
West, on average, controlling for other variables in the regression. Workers
in the Midwest earn $1.23 less per hour than workers in the West, on aver-
age, controlling for other variables in the regression. Workers in the South
earn $0.43 less than workers in the West, controlling for other variables in
the regression.
ii. The regressor W est is omitted to avoid perfect multicollinearity. If W est
is included, one of the four region’s slope coefficient would become uniden-
tifiable.
iii. The expected difference in earnings between Juanita and Jennifer is −0.43
− (−1.23) = $0.80/hour.
(e) i. The t-statistic is 8.31/0.23 = 36.1 > 1.96, so the coefficient is statistically
significant at the 5% level. The 95% confidence interval is 8.31 ± (1.96 ×
0.23).
ii. The t-statistic is −3.85/0.23 = −16.7, and 16.7 > 1.96, so the coefficient
is statistically significant at the 5% level. The 95% confidence interval is
−3.85 ± (1.96 × 0.23).
(f) i. Yes, age is an important determinant of earnings. Using a t-test, the
t-statistic is 0.51/0.04 = 12.8, implying that the coefficient on age is
statistically significant at the 5% level. The 95% confidence interval is
0.51 ± (1.96 × ×0.04).
ii. ∆Age × [0.51 ± 1.96 × 0.04] = 5 × [0.51 ± 1.96 × 0.04] = 2.55 ± 1.96 × 0.20 =
2.16 to 2.94

1
 (g) i. The F -statistic testing the coefficients on the regional regressors are zero
is 7.38. Since the 5% critical value is 2.6049. Because 7.38 > 2.6049, the
regional effects are significant at the 5% level because the null hypothesis
that the slope coefficients associated with regional dummies simultaneously
equal 0 should be rejected.
ii. A. The expected difference between Juanita and Molly is (X6,Juanita −
X6,M olly ) × β6 = β6 . Thus a 95% confidence interval is −0.43 ± (1.96 ×
0.30).
B. The expected difference between Juanita and Jennifer is

(X5,Juanita −X5,Jennif er )×β5 +(X6,Juanita −X6,Jennif er )×β6 = −β5 +β6 .

A 95% confidence interval could be constructed using the general meth-

ods discussed in Section 7.3 of textbook. In this case, an easy way to
do this is to omit Midwest from the regression and replace it with
X5 = W est. In this new regression the coefficient on South measures
the difference in wages between the South and the Midwest, and a 95%
confidence interval can be computed directly.

3. (a) Define γ = β1 − β2 . Estimate Yi = β0 + γX1i + β2 (X1i + X2i ) + ui and test

whether γ = 0.
(b) Define γ = β1 + 2β2 . Estimate Yi = β0 + γX1i + β2 (X2i − 2X1i ) + ui and then
test whether γ = 0.
(c) Define γ = (β1 + β2 ) − 1. Estimate Yi − X1i = β0 + γX1i + β2 (X2i − X1i ) + ui
and then test whether γ = 0.