Help using Base SAS procedures

proc mixed covariance structure: fixing some parameters to 0

Reply
New Contributor
Posts: 4

proc mixed covariance structure: fixing some parameters to 0

 

Hello,

 

When estimating a multilevel model using proc mixed, is there a way to fix some of the covariances in the G matrix to 0 while freely estimating the other components? Here is the matrix I want to estimate:

 

var

cov   var

cov   cov   var

  0      0      0    var

  0      0      0     0   var

  0      0      0     0    0    var

Respected Advisor
Posts: 2,655

Re: proc mixed covariance structure: fixing some parameters to 0

Assuming that you have an unstructured matrix, and you can make reasonable assumptions regarding the variances and covariances to be estimated, then the PARMS statement should be what you need:

 

PARMS (var1) (cov12) (var2) (cov13) (cov23) (var3) (0) (0) (0) (var4) (0) (0) (0) (0) (var5) (0) (0) (0) (0) (0) (var6)/hold=7,8,9,11,12,13,14,16,17,18,19,20;

 

Insert appropriate values for var1-var6 and the covariances.  OR set the variances to 1 and covariances to 0, and let the optimizer work.

 

Steve Denham

New Contributor
Posts: 4

Re: proc mixed covariance structure: fixing some parameters to 0

[ Edited ]
Posted in reply to SteveDenham

Thank you so much for helping me understand the PARMS statement. What would be the criteria for determining appropriate starting values for the variances and covariances if I don't use 1s and 0s?

Respected Advisor
Posts: 2,655

Re: proc mixed covariance structure: fixing some parameters to 0

Variances and covariances are easy to come by using PROC CORR with the COV option. These would be reasonable starting values.

 

Steve Denham

Ask a Question
Discussion stats
  • 3 replies
  • 253 views
  • 1 like
  • 2 in conversation