[R-sig-ME] Fixed effects only model with lme4

Ben Bolker bolker at ufl.edu
Sun Jun 7 01:59:56 CEST 2009

  *If* you could trust that the log-likelihoods were generated using the
same additive constants in both cases (which I don't think you can), and
*if* you could trust the straight LRT for comparisons of random effects
(see Pinheiro and Bates 2000, Scheipl, etc., for why it is really closer
to a chi-squared mixture between 0 and 1 df) then you could use logLik()
to extract the loglikelihoods of both models and
pchisq(...,df=1,lower.tail=TRUE) to get a p-value (or divide that
p-value by 2 to use a 0/1-df chi-squared mixture).

  Ben Bolker

Jeroen Ooms wrote:
> 2009/6/6 Douglas Bates <bates at stat.wisc.edu>
>> Try listing them the other way around
>> anova(fm1, fm0)
>> If the first model in the call to anova is of class "lm" then the
>> method for that class is the one chosen and that method doesn't know
>> about models created by lmer.  You must list them so that the lmer
>> model comes first.
> this still returns an error:
>> fm0 <- lm(Reaction ~ Days, sleepstudy);
>> fm1 <- lmer(Reaction ~ Days + (1|Subject), sleepstudy);
>> anova(fm0,fm1);
> Error in x$terms : $ operator not defined for this S4 class
>> anova(fm1,fm0);
> Error in FUN(X[[1L]], ...) :
>   no slot of name "call" for this object of class "lm"
> similar error if I use glm instead of lm.
> 	[[alternative HTML version deleted]]
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

Ben Bolker
Associate professor, Biology Dep't, Univ. of Florida
bolker at ufl.edu / www.zoology.ufl.edu/bolker
GPG key: www.zoology.ufl.edu/bolker/benbolker-publickey.asc

More information about the R-sig-mixed-models mailing list