Professional Documents
Culture Documents
Introduction to Econometrics
rd
(3 Updated Edition, Global Edition)
by
11.1. (a) The t-statistic for the coefficient on Schooling is 0.272/.029 = 9.38, which is
significant at the 1% level.
(d) zJed 4.107 0.272 (24) 2.421; (2.421) 0.992, this is unlikely to be
accurate because the sample did not include anyone with more than 24 years of
schooling.
11.2. (a) The t-statistic for the coefficient on Schooling is 0.551/0.062 = 8.89, which is
significant at the 1% level.
1 1
Prob Matthew ( 8.146 0.551 16)
0.662
1 e 1 e 0.67
1 1
ProbChristopher ( 8.146 0 .551 12)
0.18
1 e 1 e1.534
(b)
The shape of the regression functions is similar but the logit regression lies
above the probit function for years of schooling above 14 years, and below the
probit function for years of schooling between 8 and 14 years.
11.3. (a) The t-statistic for the coefficient on Schooling is 0.035/0.003 = 11.67, which is
significant at the 1% level.
(b)
The probabilities are very different and for years of schooling below 5 years, the
LPM model produces non-sensical results (negative probabilities).
11.4.(a)
(b) Because there is only regressor and it is binary (Male), estimates for each model
show the fraction on males and females passing the test. Thus, the results are
identical for all models.
(c) The t-statistic for the coefficient on the interaction term is −0.344/0.096 =
−3.58, which is significant at the 1% level, suggesting that years of schooling
influence whether a person is employed by government differently by gender.
11.6. (a) For a black applicant having a P/I ratio of 0.35, the probability that the
application will be denied is (2.26 2.74 0.35 0.71) (0.59)
27.76%.
(b) With the P/I ratio reduced to 0.30, the probability of being denied is (2.26
2.74 0.30 0.71) (0.73) 23.27%. The difference in denial probabilities
compared to (a) is 4.4 percentage points lower.
(c) For a white applicant having a P/I ratio of 0.35, the probability that the
application will be denied is (2.26 2.74 0.35) 9.7%. If the P/I ratio is
reduced to 0.30, the probability of being denied is (2.26 2.74 0.30)
7.5%. The difference in denial probabilities is 2.2 percentage points lower.
(d) From the results in parts (a)–(c), we can see that the marginal effect of the P/I
ratio on the probability of mortgage denial depends on race. In the probit
regression functional form, the marginal effect depends on the level of
probability which in turn depends on the race of the applicant. The coefficient
on black is statistically significant at the 1% level.
11.7. (a) For a black applicant having a P/I ratio of 0.35, the probability that the
1
application will be denied is F (4.13 5.37 0.35 1.27) 27.28%.
1 e0.9805
(b) With the P/I ratio reduced to 0.30, the probability of being denied is
1
F (4.13 5.37 0.30 1.27) 22.29%. The difference in denial
1 e1.249
probabilities compared to (a) is 4.99 percentage points lower.
(c) For a white applicant having a P/I ratio of 0.35, the probability that the
1
application will be denied is F (4.13 5.37 0.35) 9.53%. If the
1 e2.2505
P/I ratio is reduced to 0.30, the probability of being denied is
1
F (4.13 5.37 0.30) 7.45%. The difference in denial
1 e2.519
probabilities is 2.08 percentage points lower.
(d) From the results in parts (a)–(c), we can see that the marginal effect of the P/I
ratio on the probability of mortgage denial depends on race. In the logit
regression functional form, the marginal effect depends on the level of
probability which in turn depends on the race of the applicant. The coefficient
on black is statistically significant at the 1% level. The logit and probit results
are similar.
11. 8. (a) Since Yi is binary variable, we know E(Yi |Xi) 1 Pr(Yi 1|Xi) 0 Pr(Yi
0|Xi) Pr(Yi 1|Xi) 0 1Xi. Thus
Thus
var(ui | X i ) var[Yi ( 0 1 X i )i | X i ]
var(Yi | X i ) (0 1 X i )[1 ( 0 1 X i )].
Assuming that (Xi, Yi) are i.i.d., i 1,, n, the joint probability distribution of
Y1,, Yn conditional on the Xs is
n
Pr(Y1 y1 , , Yn yn | X 1 , , X n ) Pr(Yi yi |X i )
i 1
n
piyi (1 pi )1 yi
i 1
n
( 0 1 X i ) yi [1 ( 0 1 X i )]1 yi .
i 1
(b) The 95% confidence interval is 0.06 ± 1.96 × 0.021 = [0.019, 0.101].
(c) The answer in (a) will be biased if there are omitted variables, which are related
to whether an individual is self-employed as well as have an impact on
mortgage denial. Such variables would have to be related with race and standard
measures of default probability (past credit history and loan-related
information). They are included in the regressions shown in Table 9.2, so these
omitted variables are unlikely to bias the answer in (a). Other variables such as
education, marital status, and unemployment may also be related to the
probability of default, and these variables are omitted from the regression in
column. Adding these variables [see columns (4)–(6)] have a substantial effect
on the estimated effect of self-employed on the probability of mortgage denial.
11.10. (a) Let n1 (Y 1), the number of observations on the random variable Y which
equals 1; and n2 (Y 2). Then (Y 3) n n1 n2. The joint probability
distribution of Y1,, Yn is
n
Pr(Y1 y1 ,, Yn yn ) Pr(Yi yi ) p n1 q n2 (1 p q) nn1n2 .
i1
(b) The MLEs of p and q maximize the likelihood function. Let’s use the log-
likelihood function
L ln[Pr(Y1 y1 , , Yn yn )]
n1 ln p n2 ln q (n n1 n2 ) ln(1 p q).
L n1 n n1 n2
, and
p p 1 p q
L n2 n n1 n2
.
q q 1 p q
Setting these two equations equal to zero and solving the resulting equations yield
the MLE of p and q:
n1 n
pˆ , qˆ 2 .
n n
11. 11. (a) Censored or truncated regression model (note the dependent variable might
be 0).