Skip to content
Prev 283552 / 398498 Next

How do I compare 47 GLM models with 1 to 5 interactions and unique combinations?

What variables to consider adding and when to stop adding them depends greatly upon what question(s) you are trying to answer and the science behind your data.

Are you trying to create a model to predict your outcome for future predictors?  How precise of predictions are needed?

Are you trying to understand how certain predictors relate to the response? How they relate after conditioning on other predictors?

Will humans be using your equation directly? Or will it be in a black box that the computer generates predictions from but people never need to look at the details?

What is the cost (money, time, difficulty, etc.) of collecting the different predictors?

Answers to the above questions will be much more valuable in choosing the "best" model than AIC or other values (though you should still look at the results from analyses for information to combine with the other information).  R and its programmers (no matter how great and wonderful they are) cannot answer these for you.