I have a very simple regression model and I am doing the cross validation. When cv=10 the highest score i got is 60.3 and lowest is -9.7 which is useless. Average will be 30.

No of row data set= 658

enter image description here

  • $\begingroup$ What do you mean by cross-validation scores ? $\endgroup$ Commented Mar 17, 2020 at 14:23
  • $\begingroup$ I think you can try to shuffle the data before training. Maybe the data are clustered together and then training are biased on each fold $\endgroup$ Commented May 6, 2021 at 14:45

1 Answer 1


Your $R^2$ scores indicate that a linear model does not describe your data well. On top of this, there seems to be a large variability in data. You could try the following:

  • If the linear model is supposed to describe the data, check for outliers. They might be responsible for the large variation across the CV folds.
  • Try reducing the number of features if there are many. The model might be fitting noise.
  • Introducing regularization (lasso or ridge regression) might make the model more robust. This should decrease the variability of the CV errors, but the $R^2$ scores will get even worse.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.