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
- /
- Analytics
- /
- Stat Procs
- /
- A question about PROC MIXED

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

02-08-2012 01:51 PM

Hi all,

I have a question regarding PROC MIXED.

I am setting up a linear mixed model as follow:

PROC MIXED data=work.sample covtest;

class v1 v2;

model dv = / solution;

random intercept / subject=v1 solution v vcorr;

random intercept / subject=v2 solution v vcorr;

run;

v1 and v2 are categorical variables, each having more than 100 categories.

Question 1:

I assume that I have set v1 and v2 to be uncorrelated. Am I right?

Question 2:

The analysis takes me a day to complete. Is there anyway to make the syntax more efficient so that it will run faster?

Question 3:

How can I obtain the correlation between v1 and v2? Which covariance structure should I choose? (I am only interested in (1) whether v1 and v2 are correlated, and (2) the magnitude of correlation between v1 and v2.

Thank you very much for your help!

Chester