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
- /
- SAS Procedures
- /
- which procedure should i use

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

07-18-2008 01:54 PM

I have a categorical variable with 4 categories as dependent variable, and 4 categorical independent variables with 2 categories each variable and one continuous independent variable. I want to know which independent contribute the most to the variance of dependent variable.

Could anybody tell me which procedure I should use to do this kind of analysis?

Thanks

Could anybody tell me which procedure I should use to do this kind of analysis?

Thanks

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

Posted in reply to deleted_user

07-19-2008 03:40 AM

Variance of a categorical variable ? How dow you compute that ?

If likelihood can be used in your question instead of variance, I would give PROC LOGISTIC a try : the type 3 tests would tell which independent variable best contributes to model likelihood.

Regards.

Olivier

If likelihood can be used in your question instead of variance, I would give PROC LOGISTIC a try : the type 3 tests would tell which independent variable best contributes to model likelihood.

Regards.

Olivier

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

Posted in reply to deleted_user

07-21-2008 03:02 PM

Thanks a lot.

Do you think Proc Discrim can work in this case? I want to know which variables contribute the most to predict group difference. I tried with this procedure but it seemed like the procedure wouldn't work when there were categorical independent variables.

Do you think Proc Discrim can work in this case? I want to know which variables contribute the most to predict group difference. I tried with this procedure but it seemed like the procedure wouldn't work when there were categorical independent variables.