0

I've been trying to look for an answer related to my problems but wasn't able to find any. Hope you can shed some light.

I am running several algorithms to do Classification Prediction, however, when I checked my DV & IV correlation, I am getting very low result (highest is around 0.23, with majority lies in the 0.00x). From what I know, variable correlation is very important when you're applying ML algorithms such as linear and logistic regression? But I've also heard that some ML Algorithms are not affected by variable correlation?

Mind you, when I checked the accuracy of my models it leads to very high accuracy of >90% (used logistic regression, decision tree, random forest).

Thank you.

Ninja
  • 1
  • Welcome to CV. Note that your username, identicon, & a link to your user page are automatically added to every post you make, so there is no need to sign your posts. In fact, we prefer you don't. – kjetil b halvorsen Dec 13 '20 at 20:40
  • Note that if you have many predictors, each one only with a low correlation with response, they might still collectively be good, effective predictors! See for examples https://stats.stackexchange.com/questions/90711/can-independent-variables-with-low-correlation-with-dependent-variable-be-signif https://stats.stackexchange.com/questions/276974/regression-with-features-that-are-not-very-correlated-with-the-dependent-variabl – kjetil b halvorsen Dec 13 '20 at 20:49

0 Answers0