Why is CART hardly used for regression?

Is there any significant reason for its unpopularity in regression techniques?

  • $\begingroup$ Why do you say that it is hardly used for regression? $\endgroup$
    – naive
    Commented Feb 25, 2019 at 7:20
  • $\begingroup$ Because i could see many data scientist opts for lasso, ridge, elasticnet or advanced boosting model for most of the problems while CART is predominantly preferred for classification.. So i just wanna know is there any significant disadvantage of using it in regression $\endgroup$ Commented Feb 25, 2019 at 7:24

1 Answer 1


Regression and classification trees are nearly identical in how they function. I'm not aware of any specific downside to regression trees specifically. However, trees in general do have the downside of being susceptible to overfitting:

The main downside of decision trees is that even with the use of pre-pruning, they tend to overfit and provide poor generalization performance. Therefore, in most applications,...ensemble methods...are usually used in place of a single decision tree.

Muller & Guido, Introduction to Machine Learning with Python, 2017.


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.