I am using gbm model to fit a continuous dependent variable Y with several categorical variables, say, X, Z, V, and W. Suppose X has many levels (distinct values) and Y has moderate number of levels, V and W do not have so many levels. There could be some correlation among X, Z, V, and W. My question is, how would gbm estimate the influence of these variables, would it favor X with more 'explanatory power'? Does it penalize variables with too many distinct values?


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.