[R] glmmPQL, log-likelihoods issue

[Andrew Perrin]

On Fri, 21 Nov 2003, Douglas Bates wrote:

> Andrew Perrin <clists at perrin.socsci.unc.edu> writes:
>
> > Sorry for my ignorance, but could you explain a little further?  I'm
> > guessing from your response that this makes the log-likelihood that is
> > quoted by glmmPQL a poor measure of model fit. Are there are statistics
> > that would be better for reporting model fit?
>
> You could try GLMM from the lme4 package instead.  It has two methods
> ...


Does this mean anything to you?:


> morality.full.pql4<-GLMM(formula =  r.logic.morality ~ realage +
minority + female +
+ education + income + scenario + grouptype, random = ~1 |
+ groupid/participantid,
+ family = binomial(link = probit),
+ data = fgdata.df[coded.logic, ],
+ na.action = na.omit,
+ niter = 50,
+ method='PQL')
Error in lmeLevel(random[[i]], groups[[i]], reStructColumns[[i]], if
(self at useWeighted) self at weighted else self at original) :
        No direct or inherited method for function "lmeLevel" for this
call


----------------------------------------------------------------------
Andrew J Perrin - http://www.unc.edu/~aperrin
Assistant Professor of Sociology, U of North Carolina, Chapel Hill
clists at perrin.socsci.unc.edu * andrew_perrin (at) unc.edu