Conceptually, for both LASSO and ridge regression methods as lambda becomes "very large", the penalty impact grows and the coefficients approach zero. However,

a) is there a particular threshold for lambda beyond which this statement holds true?

b) Can we always infer that as lambda takes this very large value, that the coefficients will approach zero and as such, the number of predictors in the model will decrease?


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.