S&W, Chapter 10 Solutions

S&W, Chapter 10
Solution Exercises
Exercise 10.1
(a) With a $2 increase in the beer tax, the expected number of lives that would be
saved is 0.69  2 per 10,000 people holding the other regressors constant. Since
New Jersey has a population of 8.85 million, the expected number of lives saved
is 2  0.69  885 = 1221.30, holding the other regressors constant. The 99%
confidence interval for the change in death rate is [-0.69 ± 2.58  0.35]  2  885
= [-2819.6; 377].
(b) When New Jersey lowers its drinking age from 21 to 19, the expected fatality rate
decreases by 0.076 deaths per 10,000 holding the other regressors constant. The
95% confidence interval for the change in death rate is [-0.076 ± 1.96  0.068] = [-
0.20928, 0.05728]. With a population of 8.85 million, the number of fatalities is
expected to decrease by 0.076  885 = 67.26 holding the other regressors
constant with a 95% confidence interval [-185.2128, 50.6928].
(c) When real income per capita in New Jersey increases by 3%, the expected fatality
rate increases by 1.790.01 3 deaths per 10,000 holding the other regressors
constant. The 95% confidence interval for the change in death rate is [1.79 ±1.96
 0.64]  0.01  3 = [0.016068, 0.0912]. With a population of 8.85 million, the
expected number of fatalities will increase by 47.53 holding the other regressors
constant with a 95% confidence interval [14.22, 80.712].
Exercise 10.1 (ctd.)
(d) Standard errors should be clustered by groups within which the values of the
dependent variable are likely to be correlated with one another. In this case, the
natural candidate group is to cluster by state.
(e) If we believe drinking age affects the number of fatalities due to alcohol through
a non-linear function, then age should enter as a continuous variable. In this case,
the question is probably less important because both the continuous variable is
insignificant at the 10% level, and a joint F test of the three age indicators is also
insignificant at the 10% level.
Exercise 10.2
(a) For each observation, there is one and only one binary regressor equal to one.
That is,
𝐷1𝑖 + 𝐷2𝑖 + 𝐷3𝑖 = 1 = 𝑋0,𝑖𝑡
(b) For each observation, there is one and only one binary regressor equal to one.
That is,
𝐷1𝑖 + 𝐷2𝑖 + ⋯ + 𝐷𝑛𝑖 = 1 = 𝑋0,𝑖𝑡
(c) The inclusion of all the binary regressors and the “constant” regressor causes
perfect multicollinearity. The constant regressor is a perfect linear function of the
n binary regressors. OLS estimators cannot be computed in this case. Your
computer program should print out a message to this effect. (Different programs
print different messages for this problem. Why not try this, and see what your
program says?)
Exercise 10.4
(a) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0
(b) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0
(c) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0 + 𝛾3
(d) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0 + 𝛾3

Exercise 10.5
Let 𝐷2𝑖 = 1 if 𝑖 = 2 and 0 otherwise; 𝐷3𝑖 = 1 if 𝑖 = 3 and 0 otherwise … 𝐷𝑛𝑖 = 1 if
𝑖 = 𝑛 and 0 otherwise.
Let 𝐵2𝑡 = 1 if t = 2 and 0 otherwise; 𝐵3𝑡 = 1 if t = 3 and 0 otherwise … 𝐵𝑇𝑡 = 1 if

t = 𝑇 and 0 otherwise.
Let 𝛽0 = 𝛼1 + λ1
𝛾𝑛 = 𝛼𝑛 − 𝛽0
𝛿𝑇 = λ 𝑇 − 𝛽0
Exercise 10.6
Eq. 10.28  𝑣෤𝑖𝑡 = 𝑋෨𝑖𝑡 𝑢𝑖𝑡
First note that 𝐸 𝑣෤𝑖𝑡 = 𝐸 𝑋෨𝑖𝑡 𝑢𝑖𝑡 = 𝐸 𝑋෨𝑖𝑡 𝐸 𝑢𝑖𝑡 𝑋𝑖1 , 𝑋𝑖2 , … , 𝑋𝑖𝑇 , 𝛼𝑖 = 0 from
assumption 1.
Thus, 𝑐𝑜𝑣 𝑣෤𝑖𝑡 𝑣෤𝑖𝑠 = 𝐸 𝑣෤𝑖𝑡 𝑣෤𝑖𝑠 = 𝐸 𝑋෨𝑖𝑡 𝑋෨𝑖𝑠 𝑢𝑖𝑡 𝑢𝑖𝑠 .
The assumptions in Key Concept 10.3 do not imply that this term is zero. That is, Key
Concept 10.3 allows the errors for individual i to be correlated across time periods.
Exercise 10.8
There are several ways. Here is one: let 𝑌𝑖𝑡 = 𝛽0 + 𝛽1 𝑋1,𝑖𝑡 + 𝛽2 𝑡 + 𝛾2 𝐷2𝑖 + ⋯ +
𝛾𝑛 𝐷𝑛𝑖 + 𝛿2 𝐷2𝑖 × 𝑡 + ⋯ + 𝛿𝑛 𝐷𝑛𝑖 × 𝑡 + 𝑢𝑖𝑡
Where, 𝐷2𝑖 = 1 if 𝑖 = 2 and 0 otherwise and so forth.
The coefficient λ𝑖 = 𝛽2 + 𝛿𝑖

S&W, Chapter 10 Solutions

Uploaded by

Copyright:

Available Formats

S&W, Chapter 10 Solutions

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S&W, Chapter 10 Solutions

Uploaded by

Copyright:

Available Formats

S&W, Chapter 10

(b) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0

(c) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0 + 𝛾3

(d) 𝑆𝑙𝑜𝑝𝑒 = 𝛽1 𝐼𝑛𝑡𝑒𝑟𝑐𝑒𝑝𝑡 = 𝛽0 + 𝛾3

Let 𝐵2𝑡 = 1 if t = 2 and 0 otherwise; 𝐵3𝑡 = 1 if t = 3 and 0 otherwise … 𝐵𝑇𝑡 = 1 if

Where, 𝐷2𝑖 = 1 if 𝑖 = 2 and 0 otherwise and so forth.

You might also like