You observe that the R-Squared value increases as you add more variables to your regression model, but the Adjusted R-Squared value decreases. What could this imply?

  • Model is becoming more accurate; continue adding variables
  • Model is biased; change the loss function
  • Model is overfitting; remove some variables
  • Model is underfitting; add more significant variables
The observed pattern where R-Squared increases but Adjusted R-Squared decreases implies that the added variables are not contributing meaningful information. R-Squared tends to increase with more variables, but Adjusted R-Squared penalizes for unnecessary complexity. This pattern could be a sign of overfitting, and some variables might need to be removed or the selection process revisited.
Add your answer
Loading...

Leave a comment

Your email address will not be published.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