I have a small dataset with just 160 data points. When I tried ordinary linear regression on the data, I could not add more than four features without vif inflating greater than 5 (I made a stepwise forward feature selection). I decided to try an elastic net model on it. The elastic net model with all the possible features on one go gave me more features in the model? (non zero coefficients). Does this mean that all my coefficients are significant?. If I create ordinary linear regression with coefficients from the elastic net. Would it still be significant?



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.