1

I have a quite a large dataset of 75 features and one dependent variable where all are continuous. I want to do a MLR. and want to reduce the number of features. But there are many feature selection algorithm available like NCA, ILFS , InfFS, ECFS, mrmr, relieff, mutinffs, fsv, laplacian, mcfs, rfe, L0, fisher, UDFS, llcfs, cfs, fsasl, dgufs, ufsol, lasso and so on... I tried some of them where each every algorithm is giving different result. Which one would be best for Multiple Linear Regression?

Fazla Rabbi Mashrur
  • 371
  • 1
  • 3
  • 6
  • 1
    You may find the recent answer by @mung insightful https://stats.stackexchange.com/a/469980/240024, as I did. – Ryan Volpi Jun 05 '20 at 04:06

0 Answers0