[R-sig-ME] Plot a "mer" object
Luciano La Sala
lucianolasala at yahoo.com.ar
Wed Jul 14 23:19:38 CEST 2010
Hello,
I fitted a glmm on a small dataset using lme4. Data consist of dependent
variable "Egg_Volume" (continuous) and independent variables "Hatching
Order" (3 factors: first, second, third), "Year" (2 factors: 2006, 2007) and
their interaction (Hatching_Order*Year). Nest IDs were included as random
intercepts.
Model output:
> model.2 <- lmer(EggVolume~HatchOrder*Year+(1|NestID),REML=FALSE)
Linear mixed model fit by maximum likelihood
Formula: EggVolume ~ HatchOrder * Year + (1 | NestID)
AIC BIC logLik deviance REMLdev
745.4 768.4 -364.7 729.4 720.4
Random effects:
Groups Name Variance Std.Dev.
NestID (Intercept) 25.272 5.0271
Residual 5.930 2.4352
Number of obs: 130, groups: NestID, 55
Fixed effects:
Estimate Std. Error t value
(Intercept) 79.6515 1.1092 71.81
HatchOrderSecond -0.5676 0.7714 -0.74
HatchOrderThird -4.7545 0.8817 -5.39
Year2007 3.6288 1.5408 2.36
HatchOrderSecond:Year2007 -2.8466 1.0600 -2.69
HatchOrderThird:Year2007 -2.8900 1.1946 -2.42
Correlation of Fixed Effects:
(Intr) HtchOS HtchOT Yr2007 HOS:Y2
HtchOrdrScn -0.267
HtchOrdrThr -0.221 0.367
Year2007 -0.720 0.192 0.159
HtcOS:Y2007 0.195 -0.728 -0.267 -0.294
HtcOT:Y2007 0.163 -0.271 -0.738 -0.297 0.404
I've been trying to come to grips with the function "plotLMER.fnc" from
"languageR" package to plot my results, but so far I have not succeeded.
>From the relevant documentation
(http://bm2.genes.nig.ac.jp/RGM2/R_current/library/languageR/man/plotLMER.fn
c.html) there are some arguments which I am not sure how to specify
(indicated with "???" below).
plotLMER.fnc(mixto.4a, xlabel = NA, xlabs = "Year", ylabel = "Egg Volume",
ylimit = NA, fun = NA, pred = NA, n = ????, intr = ????, "end", mcmcMat =
NA, lockYlim = TRUE, addlines = TRUE, withList = FALSE, cexsize = 0.5)
Could someone help me out with this? If there is a better and simpler way of
plotting this kind of models, I'd like to know it. I am quite new to R and
its language.
Thanks in advance!
LFLS
More information about the R-sig-mixed-models
mailing list