6.1. Introduction
In the last chapter we studied the multinomial logit model for dependent variables with three or more unordered categories. When the categories are ordered, it would not be incorrect to simply ignore the ordering and estimate a multinomial model. However, there are two reasons for preferring models that take the ordering into account:
They are much easier to interpret.
Hypothesis tests are more powerful.
The disadvantage of ordered models is that they impose restrictions on the data that may be inappropriate. So whenever you use an ordered model, it’s important to test whether its restrictions are valid.
Unlike the unordered case, there are several different ways of generalizing the logit model to handle ordered categories. ...
Get Logistic Regression Using SAS®: Theory and Application now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
