Confidence interval for the effect of treatment among a certain group

[corkee]

Hi, 

I am running a logistic regression model (a relatively simple one) and want to know if it's possible to obtain a confidence interval for the effect of age among males.

This is my model: 

logit(y) = age + male + age*male

where male = 1 if male and male = 0 if female. I am not sure if I need a CONTRAST statement in PROC LOGISTIC. I know I need the variance-covariance matrix, but I am not sure how to obtain it using SAS. However, it would be ideal if I could somehow obtain the 95% CI without having to do the calculations by hand using the variance-covariance matrix.

[Reeza]

Why not use the ODDSRATIO statement?

Maybe this example helps?

http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_logistic_exa...