[R] Specify var-covar matrix in mixed linear model using lme?

From: Pengyuan Liu <pyliu8_at_yahoo.com>
Date: Sat, 1 Dec 2007 13:40:30 -0800 (PST)

I have a question about specifying var-cov matrix in mixed linear model using lme. For example, for single-level mixed linear model:
yi = XiB + Ziui + ei
ui ~ N(0,D), ei ~ N(0,sigma^2R),
var(yi)=sigma^2(ZiAZi^T + R). R is an identify matrix and A is a known var-covar matrix in my data. In my data, there is only one random effect besides ei. But this random effect is dependent among different subjects within group (this dependence is characterized in A which is known). corStruct class in nlme can specify correlation structure for within group errors (that is, specify R for ei). But I don't know how to specify A for a specific random effect. In other words, I want to fix matrix A in the analysis. Thanks a lot for your help.

Pengyuan Liu
Dept of Surgery
Washington Univ in St Louis

Be a better sports nut! Let your teams follow you

R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Sat 01 Dec 2007 - 21:47:25 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Sat 01 Dec 2007 - 22:30:16 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.