Hi, My name is José María Gómez, and I am pretty new in R. Thus, I apologize deeply if my questions are extremmely naïve.I have checked several available books and URL's, without finding any answer.
I'm trying to fit Generalized Linear Mixed Models via PQL. Below I provide the structure of my data set. Year and Plot are random variables. Fate is the binomial dependent. I have severe problems after calling glmmPQL: 1) I would like to nest Plot within Year, but I don't know how to argument it. 2) I'm not sure the analysis I fit (see below) is actually considering Year and Plot as random. 3) No way to obtain an analysis of deviance for the GLMM fit. 4) Year appears without degrees of freedom. As you see, too many problems. Thank you very much for any help. JM _____________________________________________ > summary (Ifate) Year Plot Fate Hab Peso Dist A:664 A:359 cache : 91 Oak :621 Min. :0.1903 Min. :1.362 B:574 B:427 comida:1147 Open : 94 1st Qu.:0.5515 1st Qu.:1.847 C:135 Pine :159 Median :0.6670 Median :2.137 D: 97 Rock : 23 Mean :0.6674 Mean :2.200 E:220 Shrub:341 3rd Qu.:0.7597 3rd Qu.:2.476 Max. :1.1386 Max. :3.872 > library (MASS) > library (nlme) > f1<-glmmPQL(Fate~Year+Plot+Hab+Peso+Dist, random=~1|Year/Plot, family=binomial, data=Ifate) > anova.glm(f1) Error in eval(expr, envir, enclos) : Object "Fate" not found > anova(f1) numDF denDF F-value p-value (Intercept) 1 1223 124.10634 <.0001 Year 1 0 0.00271 NaN Plot 4 3 0.96230 0.5343 Hab 4 1223 6.03162 0.0001 Peso 1 1223 4.34225 0.0374 Dist 1 1223 5.97248 0.0147 Warning messages: 1: NaNs produced in: pf(q, df1, df2, lower.tail, log.p) 2: NAs introduced by coercion ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html