Introduction to Econometrics

rd
(3 Updated Edition, Global Edition)

by

James H. Stock and Mark W. Watson

Solutions to Odd-Numbered End-of-Chapter Exercises:

Chapter 9
(This version August 17, 2014)

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 1
_____________________________________________________________________________________________________

9.1. As explained in the text, potential threats to external validity arise from differences
between the population and setting studied and the population and setting of
interest. The statistical results based on West Africa in 2000 might apply to some
parts of India in 2000 but not to the United Kingdom in 2000. In 2000, many parts
of India, particularly rural India, were plagued by high malnutrition and poor child
health. In contrast, the UK had very high standards of public health as well as
higher per capita food consumption, so the distribution of vitamins would most
likely not have a high impact on test scores. The results from West Africa in 2000
may continue to apply to the same region in 2015 but this depends on
improvements in public health and per capita consumption that may have taken
place over this time period.

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 2
_____________________________________________________________________________________________________

9.3. The key is that the selected sample contains only employed women. Consider two
women, Beth and Julie. Beth has no children; Julie has one child. Beth and Julie are
otherwise identical. Both can earn $25,000 per year in the labor market. Each must
compare the $25,000 benefit to the costs of working. For Beth, the cost of working
is forgone leisure. For Julie, it is forgone leisure and the costs (pecuniary and other)
of child care. If Beth is just on the margin between working in the labor market or
not, then Julie, who has a higher opportunity cost, will decide not to work in the
labor market. Instead, Julie will work in “home production,” caring for children,
and so forth. Thus, on average, women with children who decide to work are
women who earn higher wages in the labor market.

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 3
_____________________________________________________________________________________________________

γ 1β 0 − γ 0 β1 γ 1u − β1v
9.5. (a) Q = + .
γ 1 − β1 γ 1 − β1

β0 − γ 0 u − v
and P = + .
γ 1 − β1 γ 1 − β1

γ 1β 0 − γ 0 β1 β −γ
(b) E (Q ) = , E ( P) = 0 0
γ 1 − β1 γ 1 − β1

(c)
2 2
⎛ 1 ⎞ 2 2 ⎛ 1 ⎞
Var (Q) = ⎜ ⎟ (γ 1 σ u + β1 σ v ), Var ( P) = ⎜ ⎟ (σ u + σ v ), and
2 2 2 2

⎝ γ 1 − β1 ⎠ ⎝ γ 1 − β1 ⎠
2
⎛ 1 ⎞
Cov( P, Q) = ⎜ ⎟ (γ 1σ u + β1σ V )
2 2

⎝ γ 1 − β1 ⎠

p Cov(Q, P ) γ σ2 +βσ2 Cov( P, Q)

(d) (i) βˆ1 → βˆ0 → E (Q) − E ( P)
p
= 1 u2 12 V ,
Var ( P) σ u + σV Var ( P)

(ii) βˆ1 − β1 → σσu (2γ+1 −σβ21 ) > 0, using the fact that γ1 > 0 (supply curves slope up) and
2
p
u V

β1 < 0 (demand curves slope down).

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 4
_____________________________________________________________________________________________________

9.7. (a) False. Correlation between the dependent variable and the regressors reduce the
precision of the OLS estimator, but do not induce bias.

(b) False. If the error term exhibits heteroskedasticity, the standard errors need to be
corrected for this to produce unbiased estimates of X. If heteroskedasticity-robust
standard errors are not used, then estimates of X will be biased.

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 5
_____________________________________________________________________________________________________

9.9. Both regressions suffer from omitted variable bias so that they will not provide
reliable estimates of the causal effect of income on test scores. However, the
nonlinear regression in (8.18) fits the data well, so that it could be used for
forecasting.

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 6
_____________________________________________________________________________________________________

9.11. There are several reasons for concern. Here are a few.

Internal consistency: To the extent that price is affected by demand, there may be
simultaneous equation bias.

External consistency: The internet and introduction of “E-journals” may induce

important changes in the market for academic journals so that the results for 2000
may not be relevant for today’s market.

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 7
_____________________________________________________________________________________________________

 ( X i  X )(Yi  Y )
300

9.13. (a) ̂1  i 1

. Because all of the Xi’s are used (although some are used for the
 ( X i  X ) 2
300
i 1

wrong values of Yj), X  X , and 

n
i 1
( X i  X )2 . Also, Yi  Y  1 ( X i  X )  ui  u . Using
these expressions:

  ( X i  X )( X i  X )  ( X i  X )(ui  u )
0.8 n n n
( X i  X )2
̂1  1 i 1
 1 i0.8n 1
 i 1

  
n n n
i 1
( X i  X )2 i 1
( X i  X )2 i 1
( X i  X )2
1 0.8n 1 n 1 n
n
 i 1 ( X i  X )2 n
 i 0.8n1 ( X i  X )( X i  X ) n  i 1 ( X i  X )(ui  u )
 1  1 
1 n 1 n 1 n
n
 i 1
(Xi  X ) 2

n
 i 1
(Xi  X ) 2
 ( X i  X )2
n i 1

where n  300, and the last equality uses an ordering of the observations so that the first
240 observations ( 0.8  n) correspond to the correctly measured observations ( X i  Xi).

As is done elsewhere in the book, we interpret n  300 as a large sample, so we use the
approximation of n tending to infinity. The solution provided here thus shows that these
expressions are approximately true for n large and hold in the limit that n tends to infinity.
Each of the averages in the expression for ˆ1 have the following probability limits:

1 n p


n i 1
( X i  X ) 2
  X2 ,

1 0.8n p

n
 i 1
( X i  X ) 2
 0.8 X2 ,

1 n  p


n i 1
( X i  X )(ui  u )0 , and

1 n p


n i 0.8n 1
(   X )( X  X )  0 ,
X i i

where the last result follows because X i  Xi for the scrambled observations and Xj is
p
independent of Xi for i  j. Taken together, these results imply ̂1  0.81 .

(continued on next page)

©2015 Pearson Education, Ltd.

 Stock/Watson - Introduction to Econometrics - 3rd Updated Edition - Answers to Exercises: Chapter 9 8
_____________________________________________________________________________________________________

9.13 (continued)

p p
(b) Because ̂1  0.81 , ˆ1 / 0.8  1 , so a consistent estimator of 1 is the OLS estimator
divided by 0.8.
(c) Yes, the estimator based on the first 240 observations is better than the adjusted estimator
from part (b). Equation (4.21) in Key Concept 4.4 (page 129) implies that the estimator based
on the first 240 observations has a variance that is
1 var  ( X i   X )ui 
var( ˆ1 (240obs ))  .
 var( X i )
2
240

From part (a), the OLS estimator based on all of the observations has two sources of sampling
 i 1 ( X i  X )(ui  u ) which is the usual source that comes from the
300

error. The first is

 i 1 ( X i  X )2
300

 ( X i  X )( X i  X )
300

omitted factors (u). The second is 1 i  241

, which is the source that comes
 i 1 ( X i  X )2
300

from scrambling the data. These two terms are uncorrelated in large samples, and their
respective large-sample variances are:
  300 ( X i  X )(ui  u )  1 var  ( X i   X )ui 
var  i 1 300 
  i 1 ( X i  X )  300  var( X i )
2 2

and
  300 ( X i  X )( X i  X )  0.2
var  1 i  241 300   12 .

  i 1 i( X  X ) 2 

300

Thus

 ˆ (300obs)  1  1 var  ( X i   X )ui  2 0.2


var  1      
 var( X i )
2 1
 0.8  0.64  300 300 

which is larger than the variance of the estimator that only uses the first 240 observations.

©2015 Pearson Education, Ltd.