You have developed a regression model, and the R-Squared value is very close to 1. What could this indicate, and what would you check?

  • Good fit; No need to check anything
  • Perfect fit; Check for overfitting
  • Perfect fit; Check for underfitting
  • Poor fit; Check for bias
An R-Squared value close to 1 typically indicates a nearly perfect fit, but this might be a sign of overfitting. It is essential to verify the model's performance on unseen data, as it may be capturing noise and specificities of the training data rather than the underlying trend. Cross-validation or a hold-out validation set can help in this assessment.
Add your answer
Loading...

Leave a comment

Your email address will not be published. Required fields are marked *