Does anyone know any easy way to create an odds ratio table, analogous to a correlation table for multiple (binary) variables so that Xij would be the odds ratio for variables i and j?
William Roy
Professor of Sociology
UCLA
What does your data look like ? Post some sample data . Let me see how it work.
Check proc freq,
OR you want SAS/IML code ?
Post it at IML forum, Rick might give you something you need.
Thank you Xia. Proc freq creates one dataset per pair of variables. They can concatenated, but I would hoping for a matrix like a correlation matrix. I can do it in IML, but was hoping someone would have designed a macro to make it simple.
Bill Roy
Why would you need a data step macro ? I think IML is more simple and succinct than it . Don't you think so ?
Yes, I do. Not finding an existing macro, I've designed one myself and am working to refine it.
What does your data look like ? Post some sample data . Let me see how it work.
Xia,
I have written my own macro and no longer need help. Thank you for your offer.
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.