anova (lm, lmer ) question
On 14-10-06 12:36 PM, Farrar, David wrote:
If I understood, the following should have worked?
m.ML <- lmer(
+ log10(TWA) ~ ns(Wind.Speed,3) + isLamar + isLime1p + isLime5p + log10(TWAuw) + + (1|BiosolidSource) + (1|sample) + (1|sample.trial), + REML=F, + data=da.regr + )
m.lm <- lm(
+ log10(TWA) ~ ns(Wind.Speed,3) + isLamar + isLime1p + isLime5p + log10(TWAuw), + data=da.regr + )
anova(m.ML, m.lm)
Error in UseMethod("isREML") :
no applicable method for 'isREML' applied to an object of class "lm"
I didn't mean to imply that anova(m.ML,m.lm) would actually work, but
rather that the equivalent calculation would be appropriate, e.g.
library(lme4)
fm1 <- lmer(Reaction ~ Days + (1 | Subject), sleepstudy, REML=FALSE)
fm0 <- lm(Reaction ~ Days, sleepstudy)
NL1 <- -logLik(fm1)
NL0 <- -logLik(fm0)
devdiff <- 2*(NL0-NL1)
dfdiff <- attr(NL1,"df")-attr(NL0,"df")
pchisq(devdiff,dfdiff,lower.tail=FALSE)
The p-value is very small in this case, but that's consistent
with a large/well-determined variance estimate ...
pp <- profile(fm1)
library(lattice)
xyplot(logProf(pp))
Keep in mind that the likelihood ratio test also has some theoretical
problems in this case, mainly with boundary issues (see
http://glmm.wikidot.com/faq for more info)
-----Original Message----- From: r-sig-mixed-models-bounces at r-project.org [mailto:r-sig-mixed-models-bounces at r-project.org] On Behalf Of Ben Pelzer Sent: Saturday, October 04, 2014 10:13 AM To: r-sig-mixed-models at r-project.org Subject: Re: [R-sig-ME] anova (lm, lmer ) question Dear romunov, Ben and Ken, Thanks for your replies. From these I conclude that: - for linear (lmer vs. lm) models there's no problem in using the deviance difference - for generalized linear models (glmer vs. glm) it's ok to use the deviance difference as long as nAGQ=1. Would you agree with me? Best regards, Ben. On 4-10-2014 2:48, Ben Bolker wrote:
Thanks for checking. The comparison with Stata isn't necessarily relevant though -- or question is whether `lm` and `lmer` (or `glm` and `glmer`) include/exclude the same additive constants, so that their log-likelihoods are directly comparable. On Fri, Oct 3, 2014 at 8:38 PM, Ken Beath <ken.beath at mq.edu.au> wrote:
nAGQ=1 and greater than 1 give different results, and the nAGQ=1 matches fairly closely the log likelihood from Stata for 3 quadrature points, so presumably is correct. Stata's Laplace didn't converge with my data. Ken On 4 October 2014 09:06, Ben Bolker <bbolker at gmail.com> wrote:
romunov <romunov at ...> writes:
FWIW, this is from the glmm faq site <http://glmm.wikidot.com/faq>. How can I test whether a random effect is significant?
...
- *do not* compare lmer models with the corresponding lm fits, or
glmer/glm; the log-likelihoods are not commensurate (i.e., they
include
different additive terms)
For what it's worth, I believe this is out of date, _except_ for glmer fits with nAGQ>1. It should be possible to implement anova(<merMod>,<lm>/<glm>) -- it's only a nuisance (sadly, if we were still using S4 classes at this level it would be easier ...) Ben Bolker
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
-- *Ken Beath* Lecturer Statistics Department MACQUARIE UNIVERSITY NSW 2109, Australia Phone: +61 (0)2 9850 8516 Building E4A, room 526 http://stat.mq.edu.au/our_staff/staff_-_alphabetical/staff/beath,_ken / CRICOS Provider No 00002J This message is intended for the addressee named and m...{{dropped:11}}
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
_______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models _______________________________________________ R-sig-mixed-models at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models