5.5. Estimation with a Binary Logit Procedure
As we’ve just seen, the multinomial logit model can be interpreted as a set of binary logit equations, each equation corresponding to a comparison between two of the categories of the dependent variable. It turns out that you can legitimately estimate the multinomial logit model by running a set of binary logit models (Begg and Gray 1984). For each two-category contrast, observations that fall into other categories are excluded. Here’s how to do it for the three columns of Table 5.1:
PROC LOGISTIC DATA=my.wallet; WHERE wallet NE 2; MODEL wallet=male business punish explain; RUN; PROC LOGISTIC DATA=my.wallet; WHERE wallet NE 1; MODEL wallet=male business punish explain; RUN; PROC LOGISTIC DATA=my.wallet; ...
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.
