Dear all,
1. How do I compute the fixed-effects var-cov matrix from a GLMM object
(lme4 package) fitted with the Laplace method?
details: the vcov method for GLMM objects does not seem to work in this
case: returns a 0 * 0 matrix:
> fm1 <- GLMM(immun ~ kid2p + mom25p + ord,
+ family = binomial, data = guImmun, random = ~ 1 | comm,
method = "Laplace")
Using optimizer nlm
> fm1
Generalized Linear Mixed Model
Family: binomial family with logit link
Fixed: immun ~ kid2p + mom25p + ord
Data: guImmun
log-likelihood: -1400.842
Random effects:
Groups Name Variance Std.Dev.
comm (Intercept) 0.61621 0.78499
# of obs: 2159, groups: comm, 161
Estimated scale (compare to 1) 0.9702206
Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -0.970969 0.162467 -5.9764 2.281e-09
kid2pY 0.991837 0.119242 8.3178 < 2.2e-16
mom25pY -0.052423 0.129561 -0.4046 0.6858
ord23 -0.094408 0.140715 -0.6709 0.5023
ord46 0.040217 0.167089 0.2407 0.8098
ord7p 0.013805 0.206611 0.0668 0.9467
(Intercept) kid2pY mom25pY ord23 ord46 ord7p
[1,] -0.970969 0.9918369 -0.05242273 -0.09440838 0.04021697 0.01380546
> vcov(fm1)
<0 x 0 matrix>
#############
2. Is it relevant and efficient to use this var-cov matrix to perform
Wald-like tests for the significance of one or more fixed effects ?
version information:
# R _
platform i386-pc-mingw32
arch i386
os mingw32
system i386, mingw32
status
major 2
minor 0.0
year 2004
month 10
day 04
language R
# lme4
Package: lme4
Version: 0.6-9
Date: 2004-10-04
# Matrix
Package: Matrix
Version: 0.8-15
Date: 2004-10-02
#########
Thanks,
Renaud
--
Dr Renaud Lancelot, v??t??rinaire
C/0 Ambassade de France - SCAC
BP 834 Antananarivo 101 - Madagascar
e-mail: renaud.lancelot at cirad.fr
tel.: +261 32 40 165 53 (cell)
+261 20 22 665 36 ext. 225 (work)
+261 20 22 494 37 (home)