1

I have a GLM that I am attempting to simplify like this:

MOD.3<-update(MOD.2, ~. -WIND.SPEED)
anova(MOD.2,MOD.3,test="Chi")

When I attempt to evaluate the models with:

anova(MOD.3,MOD.2)

I get the following error:

Error in anova.glmlist(c(list(object), dotargs), dispersion = dispersion, :

models were not all fitted to the same size of dataset

I suspect this may have to do with the fact that I have missing values in the dataset, but I am not sure if this is the cause of the error nor how to approach it...

Community
  • 1
  • 1
user1658170
  • 814
  • 2
  • 14
  • 24
  • Yes, missing values are the cause of the error. Use `na.omit` to remove observations that have missing values in the subset of the data that is used in any of the models. – Roland Nov 19 '13 at 18:39

0 Answers0