Re: bivariate binomial-normal analysis

From: damianCollins <asremlforum_at_VSNI.CO.UK>
Date: Fri, 5 Feb 2010 23:19:03 +0000

Dear Jeremy,

No, sorry, it is not possible to do the bivariate binomial-normal in ASReml-R at present. It is on the todo list for the next ASReml-R release.

We would also like to add a general caution that ASReml uses the PQL technique to estimate this model. We have included this PQL technique for fitting GLMMs for the benefit of advanced users. It is well-known that PQL can give quite biased results for univariate data, as discussed in section 6.8, p112 (138) of the ASreml manual. We expect that these biases could also be non-negligible with bivariate analysis like this, especially if you have low numbers of individuals per sire (e.g. <20, say).


Dr Damian Collins, Senior Biometrician, I&I NSW
on behalf of the ASReml Project Team

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

Sent using Mail2Forum (

Read this topic online here:

-------------------- m2f --------------------
Received on Fri Feb 05 2010 - 23:19:03 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)