Binary Choice Models Flashcards
1
Q
What are the 4 scales of measurement?
A
- Nominal (Categorical)
- Ordinal
- Interval
- Ratio
2
Q
What is a nominal scale variable?
A
- Values provide classification but not an indication of order
- Numbers are only used to categorise objects
- Mathematical operations are meaningless
3
Q
What is an ordinal scale variable?
A
- Values provide an indication of order
- Cannot determine distances between the objects
- Mathematical operations are meaningless
4
Q
What is an interval scale variable?
A
- Values provide an indication of order
- Have an equal distance between scale points
- Ratios are meaningless
5
Q
What is a ratio scale variable?
A
- Values provide indication of order
- Have an equal distance between scale points
- Zero represents an absence of the object
- Mathematical operations are meaningful
6
Q
What are the problems with the Linear Probability Model?
A
7
Q
What is the log likelihood ratio index?
A
8
Q
What is the Pseudo R-Squared?
A
9
Q
How do we draw the graph to represent when y*=1 or 0 for a binary variable?
A
10
Q
What is the condition for y=1?
A
Pr(y=1│x)=Pr(ε>-[α+βx]|x)
11
Q
What is the variance of the error term in the probit model?
A
1
12
Q
What is the variance of the error term in the logit model?
A
π2/6
13
Q
How do the probit and logit models differ when normalised?
A
the logit model has a variance of √1.6
14
Q
How do we normalise the logit and probit models?
A
