I am using tree-based methods for predictions, my series are such that they are usually high but then they slowly decrease to a lower level, I am giving the month and exogenous things as feature, but how would machine learning differentiate between pattern after some year, for example, pattern cj=hange occur after 2019. giving year as a feature in tree-based models runs into a problem also, when I predict new year, tree-based model would not have been trained on that new year so the problem would still persist.


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.