Is the labeling/naming of levels in the documentation for the
predict.glmmPQL function "backwards"? The documentation states "Level
values increase from outermost to innermost grouping, with level zero
corresponding to the population predictions". Taking the sample in
the documentation:
fit <- glmmPQL(y ~ trt + I(week > 2), random = ~1 | ID,
family = binomial, data = bacteria)
> head(predict(fit, bacteria, level = 0, type="response"))
[1] 0.9680779 0.9680779 0.8587270 0.8587270 0.9344832 0.9344832
> head(predict(fit, bacteria, level = 1, type="response"))
X01 X01 X01 X01 X02 X02
0.9828449 0.9828449 0.9198935 0.9198935 0.9050782 0.9050782
> head(predict(fit, bacteria, type="response")) ## population prediction
X01 X01 X01 X01 X02 X02
0.9828449 0.9828449 0.9198935 0.9198935 0.9050782 0.9050782
The returned values for level=1 and level=<unspecified> match, which
is not what I expected based upon the documentation. Exponentiating
the intercept coefficients from the fitted regression, the level=0
values match when the random effect intercept is included
> 1/(1+exp(-3.412014)) ## only the fixed effect
[1] 0.9680779
> 1/(1+exp(-1*(3.412014+0.63614382))) ## fixed and random effect intercepts
[1] 0.9828449
Thanks!
Mike
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
