As important as I have found external model validation to be, there is certainly a lack of material out there. The closest thing I have found is a paper that is focused on external validation for a logistic regression model.
External validation in this context refers to applying a model to a different population than the model was originally built on.
I am trying to build up a defense for a piece of intuition I have about this topic. I have a SVM regression model that is build on ~ 15K observations. The modeling process was a difficult one. My intuition tells me that the dataset for this external validation should be as large as the dataset the model was built on (15K). Does this intuition line up with any work that has been done?
