Re: Multivariate analysis using a giv file

From: Arthur <asremlforum_at_VSNI.CO.UK>
Date: Fri, 16 Jan 2009 03:04:21 +0000

Dear Yutao Li,

The first thing I note is that 'animal' is coded 1:1097 (576 unique values) in the data file but is just declared as a covariate (not a factor).

What ASReml expects when it sees giv(animal,1) is that animal is a factor
and that the GIV matrix matches it.

So, declare animal as
 animal !I 576

The identities in the data file appear to be in order.
ASReml assumes the GIV file is coded in the same order.

Now parts 1 and 2 should run.

Part 3 needs revision:
!part 3
weight ~ mu environ !r environ.animal

#There are two separate errors, with one dimension
2 1 1
#error environ 1
480 0 IDEN !S2=80
96 0 IDEN !S2=65
environ.animal 2 # Remove # from start of line
environ 0 CORR 0.65 15 20 # Prefer CORUH rather than CORR
animal 0 GIV1 # Need to specify the variance structure here

Hopefully that will sort it.


Arthur Gilmour

Principal Research Scientist (Biometrics)
NSW Dept Primary Industries

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

Sent using Mail2Forum (

Read this topic online here:

-------------------- m2f --------------------
Received on Sat Jan 16 2009 - 03:04:21 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.)