Problem Set #3 Solutions (1) The intercept tells you the expected birthweight of a child whose mother never smoked while pregnant. Thus, E(BirthW eight|Cigs = 0) = 119.6. Since there 16 ounces per pound, this suggests that the average birthweight for moth
Solutions: Problem Set #6 (1) The coefficient estimates for the regressions in this problem set are provided in the other file. These are the same as reported in the textbook. (1a) In class, we derived the omitted variables bias formula: 2 2 + 3 Cov(x, z)
Solutions: Problem Set #1
(1) The following table gives the joint probability distribution p(X, Y ) of random variables X and Y .
Y 1 2 3 4
1 .02 .03 .04 .09
X 2 .04 .18 .04 .18
3 .08 .04 .08 .18
Determine the following:
(a) Do the entries of the table sa
Solutions: Problem Set #4 (1) Taking the derivative of the objective function with respect to 1 gives the first order condition: -2
i=1 n
^ (yi - 1 ) = 0.
Cleaning this expression up a bit, we obtain
n i=1
^ yi - n1 = 0
which implies
1 n ^ 1 = y yi . n i=
Solutions: Problem Set #10 (1a) Mean-independence could be violated for a variety of reasons. First, highlymotivated students may be more likely to purchase computers. (They decide to purchase a computer, simply because they think it will give them an "ed
Solutions: Problem Set #8 (1) The regression output is provided in the other le. Increasing team batting average by, say, 10 points will lead to about 5 more wins per season. Hitting 10 more home runs in a season will lead to about 1.2 more wins in that s
Solutions: Problem Set #5
(1) (a) Note
E (W age|M ale = 1) = 12.68 + 2.79 = 15.47
and
E (W age|M ale = 0) = 12.68.
Thus, the gender gap is
E (W age|M ale = 1) E (W age|M ale = 0) = 2.79.
(b) First, the t-statistic associated with the null hypothesis that
Solutions: Problem Set #7
(1) The regression output and STATA code are provided in the other le.
(1b) To implement this test, we need to get the RSS values from both the restricted
and unrestricted models. Note that we should not use the R-squared version