In my machine learning class, we have learned about how LASSO regression is very good at performing feature selection, since it makes use of $l_1$ regularization.

My question: do people normally use the LASSO model just for doing feature selection (and then proceed to dump those features into a different machine learning model), or do they typically use LASSO to perform both the feature selection and the actual regression?

For example, suppose that you want to do ridge regression, but you believe that many of your features are not very good. Would it be wise to run LASSO, take only the features that are not near-zeroed out by the algorithm, and then use only those in dumping your data into a ridge regression model? This way, you get the benefit of $l_1$ regularization for performing feature selection, but also the benefit of $l_2$ regularization for reducing overfitting. (I know that this basically amounts to Elastic Net Regression, but it seems like you don't need to have both the $l_1$ and $l_2$ terms in the final regression objective function.)

Aside from regression, is this a wise strategy when performing classification tasks (using SVMs, neural networks, random forests, etc.)?

  • 2
    $\begingroup$ Yes, Using lasso for feature selection for other models is a good idea. Alternatively tree based feature selection could also be fed to other models $\endgroup$ May 9, 2016 at 23:31
  • 1
    $\begingroup$ The lasso only performs features selection in linear models -- it doesn't test for higher-order interactions or nonlinearity in the predictors. For an example of how that might be important: stats.stackexchange.com/questions/164048/… Your mileage may vary. $\endgroup$
    – Sycorax
    May 10, 2016 at 1:02

2 Answers 2


Almost any approach that does some form of model selection and then does further analyses as if no model selection had previously happened typically has poor properties. Unless there are compelling theoretical arguments backed up by evidence from e.g. extensive simulation studies for realistic sample sizes and feature versus sample size ratios to show that this is an exception, it is likely that such an approach will have unsatisfactory properties. I am not aware of any such positive evidence for this approach, but perhaps someone else is. Given that there are reasonable alternatives that achieve all desired goals (e.g. the elastic net), it this approach is hard to justify using such a suspect ad-hoc approach instead.

  • 3
    $\begingroup$ agreed.... the point is everything has to fit within a crossvalidation framework... so you should do some nested cross validation to do the two separate regularisations (otherwise you will run into problems), and nested crossvalidation is using less data for each part. $\endgroup$
    – seanv507
    May 10, 2016 at 16:56

Besides all the answers above: It is possible to calculate an exact chi2 permutation test for 2x2 and rxc tables. Instead of comparing our observed value of the chi-square statistic to an asymptotic chi-square distribution we need to compare it to the exact permutation distribution. We need to permute our data in all possible ways keeping the row and column margins constant. For each permuted dataed set we caluclated the chi2 statistics . We then compare our observed chi2 with the (sorted) chi2 statistics The ranking of the real test statistic among the permuted chi2 test statistics gives a p-value.

  • $\begingroup$ Could you add detail to your answer, please? In its current form, it is not clear how one would calculate the exact chi2 test. $\endgroup$ Aug 11, 2016 at 15:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.