turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- SAS Programming
- /
- General Programming
- /
- GLIMMIX random effects solution

Topic Options

- RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

06-16-2015 08:37 AM

Hi

I am running a logistic random effects model using GLIMMIX to predict the probability of death. I'm interested in variation between geographic areas. My first model is a "null" model with no fixed effects and a random intercept for geographic areas. The second model controls for age and sex by including those variables in the model.

model 1:

proc glimmix data=data;

class geo;

model death (event-last) = / solution dist=binary link=logit oddsratio;

random geo / solution;

run;

model 2:

proc glimmix data=data;

class geo;

model death (event-last) = age female/ solution dist=binary link=logit oddsratio;

random geo / solution;

run;

I would like to get the probability of death in each geographic area relative to the average. The coefficients in the "solution for random effects" table are on the logit scale. How do i get the probability of each area being higher, lower than 1 for each model?

Thanks very much

LNL

- Mark as New
- Bookmark
- Subscribe
- RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

06-16-2015 12:30 PM

If there are not too many levels to geo, consider using ESTIMATE statements with the ILINK option.

Steve Denham