[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

*To*: asreml@chiswick.anprod.csiro.au*Subject*: Specifying variance components*From*: bsouthey@iastate.edu*Date*: Tue, 31 Aug 1999 14:12:40 CDT*Sender*: asreml-owner@ram.chiswick.anprod.csiro.au

Hi, I need some help with specifying variance components under a multivariate model with maternal effects. The data set does not have sufficient information to estimate the additive-maternal covariance. So I am just fitting dam as a second effect. Below is a my current .as file (simplified as the random effects are the important part at present), is this a correct specification? Is this the same as saying: wt0 wtw ~ Trait !r Tr.tag at(Tr,1).dam at(Tr,2).dam ? If I decide to fit only a maternal effect to wt0, the model changes to: wt0 wtw ~ Trait !r Tr.tag at(Tr,1).dam How would I specify starting variance components for the dam component? Thanks in advance for any comments, Bruce Southey Bivariate with maternal tag !p # 2525 animals in pedigree sire dam !p br 6 !I yb 6 !I sx 2 !I aod 4 !I bd !-12260 aw ped.file wt.dat !MAXIT 1 # !CONTINUE wt0 wtw ~ Trait !r Tr.tag Tr.dam 1 2 2 # !step 0.0001 0 0 0 Trait 0 US 0.29202 0.63988 3.0156 Tr.tag 2 Trait 0 US 0.28609 0.85243 4.6901 tag 0 0 Tr.dam 2 2 0 DIAG 0.2 4.2 dam 0 0 -- Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml

- Prev by Date:
**RE: question related to the TAG paper** - Next by Date:
**Re: Specifying variance components** - Prev by thread:
**weighted least squares** - Next by thread:
**Re: Specifying variance components** - Index(es):