I am studying the elastic net regression and in some material I read, it was mentioned that the method will choose a group of regressors that are correlated while LASSO can pick one among the correlated variables.

My question is, if I apply elastic net, then can I still interpret the coefficients on the regressor to the dependent variable? Since multicollinearity may cause the coefficients to have opposite signs, I am concerned that I will not be able to interpret my coefficients.

  • 1
    $\begingroup$ I took the liberty of editing the question attempting to make it clearer. Please see if I did not change the meaning. I do not think the elastic net coefficients will suffer much from multicollinearity -- because regularization (such as elastic net or ridge regression) actually should be a sensible remedy for multicollinearity. However, I am not really qualified to answer the question. $\endgroup$ – Richard Hardy Nov 30 '15 at 20:21
  • $\begingroup$ Thanks for the editing! It's much better now! Hope some one can answer this question! :) $\endgroup$ – phebe wu Dec 2 '15 at 20:42

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.