Validating a model

[noetsi]

What is the best, or maybe reasonable, way to determine if your regression model fits the data well? I guess you could use part of the data to do this, but I would prefer not to use a lot of my data that way. I have read suggestions of generating bootstraps to do this, but not enough detail to know how to do it.

Is there a good test for overfitting?

 

[hlsmith]

Data splits and/or shrinkage (LASSO, ridge, enet, Bayes).

 

[Dason]

What is your sample size

 

[noetsi]

It was a theoretical question, but generally at least several thousand cases. Up to 40,000 sometimes.