[R-sig-ME] MCMCglmm with datasets of different lengths

[Ingleby, Fiona]

Thanks for the reply, David. I don't think I explained myself clearly enough as I don't think the model with the five traits as response variables would give me what I'm looking for. I want to include the size measurements as a covariate, something along the lines of the second model I attempted to fit (although obviously this model was incorrect) with the basic form

cbind(trait1, trait2,trait3,trait4) ~ size:sex:trait-1, random=~us(sex:trait):family, rcov=~us(trait):family

I'm hoping to be able to extract the variance-covariance matrix at the family level for each of the four traits in each sex (this is why I have tried to fit the random terms as above), as well as get an estimate for the relationship between size and each of the four traits in each sex (and so that is why I've fitted the fixed effects as above). Like I say, I don't actually know if this is possible in MCMCglmm, but I just thought I would check.

Thanks again,

Fiona


Dr Fiona C Ingleby
Postdoctoral Research Fellow 
University of Sussex
Email: F.Ingleby at sussex.ac.uk
Website: fionaingleby.weebly.com


On 1 May 2013, at 23:53, David Duffy <David.Duffy at qimr.edu.au> wrote:

On Wed, 1 May 2013, Ingleby, Fiona wrote:

I started with one dataset with four traits (trait1, trait2, trait3 and trait4) measured in males and females from 30 families.
[...]
This is fine. However, I have additional data that I'd like to use to fit a more complicated model if possible. I have size data for males and females from these families, but these data are not from the same individuals as the original data.

Generally, as long as you have the relationships of individuals between the two sets, then the model should be identified (eg estimate genetic correlation between traits expressed only in males and females).  So maybe try the five trait model directly..

Just 2c, David Duffy,

| David Duffy (MBBS PhD)                                         ,-_|\
| email: davidD at qimr.edu.au  ph: INT+61+7+3362-0217 fax: -0101  /     *
| Epidemiology Unit, Queensland Institute of Medical Research   \_,-._/
| 300 Herston Rd, Brisbane, Queensland 4029, Australia  GPG 4D0B994A v