How to generate predictive values when using Anova (Type III) on mixed effects linear model

by Jane Lim   Last Updated August 27, 2018 17:19 PM

I am running a mixed effects linear model on AB test and using Anova, Type III to determine significance. I am doing this because I have an interaction variable.

My R code is :

model = lme(DV ~ variation_id + time + variation_id*time, random = ~1|user_id, data=data, method="REML")
sstable <- Anova(model, type = 3)

Ordinarily, I would use summary(model) to get coefficients which I use to generate predictive output (DV) but I'm unable to do this with Anova.

What's the best practice for generating predictive estimates when testing significance with Anova?



Related Questions


Choosing Interactions

Updated March 27, 2019 08:19 AM


Can a permutation test be used to evaluate a model?

Updated December 07, 2018 10:19 AM


Evaluating prediction error on given deciles

Updated August 14, 2017 12:19 PM