Regression with a Binary Dependent Variable SW Chapter 11

1
Regression with a Binary Dependent Variable (SW
Chapter 11)
2
Example Mortgage denial and raceThe Boston Fed
HMDA data set
3
The Linear Probability Model(SW Section 11.1)
4
The linear probability model, ctd.
5
The linear probability model, ctd.
6
Example linear probability model, HMDA data
7
Linear probability model HMDA data, ctd.
8
Linear probability model HMDA data, ctd
9
The linear probability model Summary
10
Probit and Logit Regression(SW Section 11.2)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
Probit regression, ctd.
15
STATA Example HMDA data
16
STATA Example HMDA data, ctd.
17
Probit regression with multiple regressors
18
STATA Example HMDA data
19
STATA Example, ctd. predicted probit
probabilities
20
STATA Example, ctd.
21
Logit Regression
22
Logit regression, ctd.
23
STATA Example HMDA data
24
Predicted probabilities from estimated probit and
logit models usually are (as usual) very close in
this application.
25
Example for class discussion
26
(No Transcript)
27
(No Transcript)
28
Hezbollah militants example, ctd.
29
Predicted change in probabilities
30
Estimation and Inference in Probit (and Logit)
Models (SW Section 11.3)
31
Probit estimation by nonlinear least squares
32
Probit estimation by maximum likelihood
33
Special case the probit MLE with no X
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
The MLE in the no-X case (Bernoulli
distribution), ctd.
38
The MLE in the no-X case (Bernoulli
distribution), ctd
39
The probit likelihood with one X
40
The probit likelihood function
41
The Probit MLE, ctd.
42
The logit likelihood with one X
43
Measures of fit for logit and probit
44
Application to the Boston HMDA Data (SW Section
11.4)
45
The HMDA Data Set
46
The loan officers decision
47
Regression specifications
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
Table 11.2, ctd.
52
Table 11.2, ctd.
53
Summary of Empirical Results
54
Remaining threats to internal, external validity
55
Summary(SW Section 11.5)