[R-sig-ME] Random covariates with common variance

Juan Pedro Steibel steibelj at msu.edu
Thu Feb 19 14:38:41 CET 2009


Hello everyone,
I can not find in the documentation how to fit the following model with 
lmer, any help would be much appreciated:
Suppose I have a set of three covariates z1, z2, z3 that I want to fit 
as random effects. Their coefficients U=(u1,u2, u3) have distribution: 
U~N(0,sig.sq_u*I)
Where sig.sq_u is a scalar, I is the Identity matrix of order 3.
I can fit a model with an unstructured covariance matrix on U, and with 
independent heteroskedastic distributions on the components of U, but 
cannot find a way of fitting what I need.

Any pointers/suggestion/comments?
Thanks!
JP


-- 
=============================
Juan Pedro Steibel

Assistant Professor
Statistical Genetics and Genomics

Department of Animal Science & 
Department of Fisheries and Wildlife

Michigan State University
1205-I Anthony Hall
East Lansing, MI
48824 USA 

Phone: 1-517-353-5102
E-mail: steibelj at msu.edu




More information about the R-sig-mixed-models mailing list