Final Exam Flashcards
1
Q
When scaling from ft to miles for the variable for B1 coefficient, what is w2? What is the effect?
A
1/5280. 1/1/5280 = 5280 times larger
2
Q
Why use logit? Why not normal linear reg?
A
Range can be greater than one in linear regression. More realistic probability. No need homoscedasticity in logit (diff variance).
3
Q
What differs in interpretation for logit, probit, and LPM?
A
logit/probit has diff probability for different individuals (reference point). LPM uses OLS, logit/probit uses maximum likelihood
4
Q
A
5
Q
A
5
Q
A
6
Q
A
6
Q
A
6
Q
A
6
Q
A
7
Q
A
7
Q
A
7
Q
A
7
Q
A
8
Q
A
