Re: asreml-r bivariate glmm

Dear Michael,

GLMM models have only been developed for the case of a single GLMM trait.

It is difficult to conceive what is the appropriate error structure for bivariate GLMM traits when the variances are defined as functions of the mean.

Now one possible approach is to string out the data as a single
resonse vector and then model the error covariance as 'subject'
variance component. This is in fact how ASReml handles the multivariate model anyway but the Normal model is easier because the variance is not dependent on the mean. In standalone ASReml, this would sssimply mean using the !ASUV
qualifier and defining the appropriate model and structures.

I trust this helps.

------------------------
Arthur Gilmour

Principal Research Scientist (Biometrics)
NSW Dept Primary Industries

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=431#431

-------------------- m2f --------------------
Received on Wed Mar 24 2009 - 04:29:33 EST