Hello All,
I am having a bit of trouble determining the significance of a term in
a phylogenetic GLS using ape and nlme. I specified a model with a
continuous normal dependent variable and a single categorical factor
with 3 unranked levels as the independent variable. The model is quite
Hi Patrick,
Try the option method=ML in gls(). By default REML is used. Also you
should use anova() to compute the LRT; I guess lrtest is a function of
yours. ?anova.gls says:
Note:
Likelihood comparisons are not meaningful for objects fit using
restricted maximum likelihood and