15.12. Alternating Logistic Regression
There are some limitations of the correlation approach to fit models to binary data. The data influences the range of the correlation since the estimates of rjk are constrained by the means, μjk = Pr{yij = 1}. Consider:
The odds ratio appears to be a more natural choice for modeling the association in binary data as they are not constrained by the means.
In GEE, the correlations are treated as nuisance parameters, and the use of correlations versus odds ratios usually has little influence on inference on ...
Get Categorical Data Analysis Using The SAS® System, 2nd Edition 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.
