Apologies, but it is not currently possible to directly
access the covariate mean used within REML analysis.
It is possible to set this to your own value by adjusting
the covariate prior to analysis using CALCULATE, then
setting CADJUST=no within the analysis.
It should be possible to indirectly access the value
by saving the design matrix (single column) for the
covariate, and comparing this with the raw values, ie.
VKEEP x; DESIGN=Dx
VARIATE vDx; val=Dx
CALC [p=s] covmean=x-vDx
Would you want to see the covariate mean values printed
out in the model summary, or just to be able to save
them via VKEEP?
Sue Welham
Biomathematics Unit
Agriculture and the Environment Division
Rothamsted Research
Harpenden UK AL5 2JQ
E-mail: [log in to unmask]
Tel: +44 (0)1582 763133 ext 2278
Fax: +44 (0)1582 467116
Rothamsted Research is a company limited by guarantee, registered in
England under the registration number 2393175 and a not for profit
charity number 802038.
> -----Original Message-----
> From: J.S. Gilchrist [mailto:[log in to unmask]]
> Sent: 11 June 2004 00:09
> To: [log in to unmask]
> Subject: covariate means in REML
>
>
> Dear Genstatters,
>
> When running mixed models (REML and IRREML), the model output
> can be set to
> covariate centred to mean or to zero (CADJUST=mean,
> CADJUST=none). I wish
> to know WHAT the covariate means are for a mixed model with 2
> variates. I
> know how to do this for IRREML (using ITER="" to then
> calculate the mean
> used in the model). Can anyone tell me how to get the
> covariate means out
> of a REML model?
>
> Thanks.
>
> Jason Gilchrist
> [log in to unmask]
>
|