[R] Re: stepwise REML

[David Duffy]

Ronaldo Reis Jr. wrote:

> I know that is not possible make a stepwise procedure using REML in R, I can 
> use ML for this.
> For nested design it may be very dangerous due the difference in variance 
> structure, mainly in a splitplot design. ML make significative variables that 
> REML dont make.
> I read an article that is made a stepwise procedure using GENSTAT.
> All are made using REML.

One can always compare models in a hierarchical type fashion (providing
they can be nested).  The LRTS comparing models fitted via
REML may not be chi-square.

-- 
| 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