I understand the ins and outs of the processes of both cross validation (partition the data set evenly, train on k-1 partitions, blah blah blah) and bagging (train M models composed of n observations picked at random with replacement, blah blah blah).

It seems to me that at the most basic level they're just sampling the same training set (albeit by different methods), but cross validation is used to get a more accurate estimate of model performance on unseen test data, and bagging is used to get a more accurate fit of the training data.

If this true (someone may have to straighten me out if I have the ideas confused), does that mean you may want to use cross validation and bagging together - to find an accurate model that more accurately evaluates model performance on unseen data? It seems a little over kill, but I'm not sure.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.