anova.mppm: likelihood ratio test does not work for random effects models for replicated patterns.

[marcelinodelacruz]

Including a random term in the models of the example of anova.mppm() makes the likelihood ratio test to fail, e.g.:

H <- hyperframe(X=waterstriders)
#' test for loglinear trend in x coordinate
mod0 <- mppm(X~1, data=H, Poisson(), random = ~1|id)
modx <- mppm(X~x, data=H, Poisson(), random = ~1|id)
anova(mod0, modx, test="Chi")

Error in anova.lme(structure(list(modelStruct = structure(list(reStruct = structure(list( :
objects must inherit from classes "gls", "gnls", "lm", "lmList", "lme", "nlme", "nlsList", or "nls"
Error in anova.mppm(mod0, modx, test = "Chi") : anova failed

[baddstats]

This has arisen because recent versions of nlme have imposed stricter restrictions on the arguments of anova.lme. We are trying to find a work-around.

[baddstats]

As a temporary solution, if you set spatstat.options(developer=TRUE), your example code will work, in the development version spatstat 1.64-1.027.

[marcelinodelacruz]

Thank you so much Adrian for dealing with this. Your temporary solution works fine also with other data and examples.

[baddstats]

This solution has been adopted permanently in spatstat 1.64-1.052 and later.

So you do not need to typespatstat.options(developer=TRUE) any more.