Scribd
Upload
20 views

Ordered Probit and Logit Models

The document discusses ordered probit and logit models, which are used for ordered dependent variables like rating scales. It explains how the models work, providing equations and discussing interpretation of coefficients and marginal effects. Thresholds divide the latent variable into categories for the observed dependent variable.

Uploaded by

tegarpradana
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
20 views

Ordered Probit and Logit Models

The document discusses ordered probit and logit models, which are used for ordered dependent variables like rating scales. It explains how the models work, providing equations and discussing interpretation of coefficients and marginal effects. Thresholds divide the latent variable into categories for the observed dependent variable.

Uploaded by

tegarpradana
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Ordered Probit and Logit Models

Ani Katchova

2013 by Ani Katchova. All rights reserved.

Ordered Probit and Logit Models Overview

Ordered outcome dependent variable


Ordered probit and logit models (coefficients and marginal effects)

Ordered Probit and Logit Models

Ordered outcome examples

Rating systems (poor, fair, good, excellent)


Opinion surveys (strongly agree, agree, neutral, disagree, strongly disagree).
Employment (unemployed, part time, full time)
Ranking (senior, junior, sophomore, freshman)
Grades (A, B, C, D, E)
Bond ratings (AAA, AA, A, B, etc.)

Ordered outcome dependent variable


The categories for the dependent variables are rankings so the numbers do not make sense
(even if they are coded as 0, 1, 2, 3, 4, the difference between the first and second outcome
may not be the same as between the second and third).

Ordered logit and probit models


An index model for a single latent variable y* (which is unobservable, we only know when it
crosses thresholds).

if

The probability that observation i will select alternative j is:

For the ordered logit, F is the logistic cdf


For ordered probit, F is the standard normal cdf.

/ 1

The ordered logit/probit model with j alternatives will have one set of coefficients with (j-1)
intercepts. You can recognize an ordered choice model by the multiple intercepts.
The ordered logit/probit model with j alternatives will have j sets of marginal effects.
4

Interpretation of coefficients: the sign of parameters shows whether the latent variable y*
increases with the regressor. The magnitude of the coefficients will be different by a scale
factor between the probit and logit models.
Marginal effects for the ordered logit/probit models
The marginal effect of an increase in a regressor xr on the probability of selecting alternative j
is:

The marginal effects of each variable on the different alternatives sum up to zero.
Marginal effects interpretation: each unit increase in the independent variable
increases/decreases the probability of selecting alternative j by the marginal effect expressed as
a percent.

You might also like