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
- /
- To estimate Restricted maximum likelihood estimati...

Topic Options

- Subscribe to 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
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

08-05-2013 02:16 PM

All,

I am doing a simulation with repeated model. How can I estimate REML for regression parameters using SAS IML? Do I have to write a program? Any help I appreciate.

Thank you in advanced.

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

08-06-2013 09:54 AM

Yes, you would have to write a program to do this in IML. However, there are several procedures (MIXED, GLIMMIX, HPMIXED) that are available to get REML estimates, both fixed and random effects.

Steve Denham

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

08-06-2013 12:07 PM

Thank you so much Steve...If anybody can give me a link or any help It will great appreciate..I am new to SA, no idea how to start this..

Thank you.

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

08-07-2013 01:56 PM

I would look through the documentation for PROC MIXED, and examine the many examples. Here is a link:

Steve Denham

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

08-09-2013 01:25 PM

Thank you Steve....

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

05-14-2014 04:18 PM

As Steve noted you can use both the mixed and the glimmix procedure. MIXED has somewhat great flexibility in the types of V covariance matrices you can fit while GLIMMIX allows for non-normal distributions and has some additional options for the MCPs. They also vary in the sort of iterative process they use to estimate the covariance parameters in the V matrix.