I have three regression binomial models in R (GLMM) for which I have the following anova table.


Now, which is the model of the best fit based on the metrics in the table? Is it model 2 since it has the lowest BIC value?

I read somewhere that the best fitting model should have a lower residual deviance, AIC and BIC values than other models. The p-value threshold here is p<0.5.

In addition, how can I calculate the Chisq,Chi Df and p-value of model2 in R since these are not provided in the first row?

  • $\begingroup$ The Chisq values are for differences with the previous row. There is no previous row for model2 so no Chisq. $\endgroup$ – mdewey Sep 1 '16 at 12:30
  • $\begingroup$ Ok but still would model2 be the model of the best fit? $\endgroup$ – Shad Sep 2 '16 at 16:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.