Hi, lme() allows the use of variance function in the "weights" option. In lmer() the "weights" seems to expect actual weights rather than a function (as in lm()). Can variance functions be used in lmer()? Thanks, Michael
variance function in lmer
3 messages · Michael, Douglas Bates
On Mon, Jun 22, 2009 at 3:45 PM, Michael Li<wuolong at gmail.com> wrote:
Hi, lme() allows the use of variance function in the "weights" option. ?In lmer() the "weights" seems to expect actual weights rather than a function (as in lm()).
And it's even documented that way too :-)
Can variance functions be used in lmer()?
Not currently, no.
On Jun 22, 2009, at 5:15 PM, Douglas Bates <bates at stat.wisc.edu> wrote:
On Mon, Jun 22, 2009 at 3:45 PM, Michael Li<wuolong at gmail.com> wrote:
Can variance functions be used in lmer()?
Not currently, no.
Thanks. I thought that was the case. Since nlme/lme4 are such important packages, Is there a "roadmap" in terms how they will evolve in the near future and what features are planned (e.g., empirical stderr)? Michael