Lecture 7: Limited Dependent Variables and Maximum Likelihood

Question 1

What is the Likelihood Function

Answer 1

Question 2

What is the Log likelihood function?

Answer 2

Question 3

What is the Maximum Likelihood estimator?

Answer 3

Question 4

What are the two different types of discrete variables?

Answer 4

Question 5

What are the three different types of binary models?

Answer 5

Question 6

Describe the linear probability model

Answer 6

Question 7

What are the three advantages of the linear probability model

Answer 7

Question 8

What are the 2 disadvantages of the linear probability model

Answer 8

Question 9

What are the solutions to the disadvantahes of the linear probability model?

Answer 9

Question 10

Explain the probit model

Answer 10

Question 11

Explain the Logit model

Answer 11

Question 12

Explain marginal effects for binary models. How does it differ for linear vs non-linear models?

Answer 12

Question 13

What is the odds ratio and the log-odds ratio?

Answer 13

Question 14

What are the two methods for looking at goodness of fit for a nonlinear model?

Answer 14

Question 15

Explain the 4 steps for the Percentage correctlu predicted Gooness of fit test

Answer 15

Question 16

Explain the Pseudo R-squred goodness-of-fit test

Answer 16

Question 17

Should we use LPM or probit/logit models?

Answer 17

Question 18

Explain what censored models are

Answer 18

Question 19

Explain what Truncated models are

Answer 19

Question 20

What happens if we do OLS estimation on censored or tuncated data?

Answer 20

Question 21

How does Maximum likelihood (ML) estimation address censoring and tuncation?

Answer 21

Question 22

How does censored ML esatimation work?

Answer 22

Question 23

What are selection models?

Answer 23

Question 24

Describe the set up on the selction problem for wages

Answer 24

Question 25

Explain the Heckman 2 stage selection model?

Answer 25

Question 26

What exclusion restrictions are required for the Heckman selection model?

Answer 26