CODEWITHSUNDEEP
pythonmachine-learningregressionlinear-regressionrandom-forest
Yoshitha Penaganti
Yoshitha Penaganti

Reputation: 464

Negative r squared for linear regression but positive r squared with Random forest regressor

My R squared score for Linear regression is -1.56 but my R squared score for Random Forest regressor is around 0.48. Is it okay to get scores like this?

Upvotes: 0

Views: 521

Answers (1)

paradocslover
paradocslover

Reputation: 3294

Is it okay to get scores like this?

Yes.

Explanation - Linear Regression and Random Forest are two very different models. It is highly possible that the data on which you are trying to fit the model is so complicated that linear regression does not fit it well.

Regarding the negative R squared score - It simply means that the linear regression model fits worse than a horizontal line.

Upvotes: 1

Related Questions