I have a analysis that is looking to predict the total customer value based on a customer's first purchase amount. I am noticing that a set of features predict whether the customer will purchase ever again after their first purchase (first purchase = total value) and a totally separate set of features where, GIVEN A CUSTOMER WILL PURCHASE AGAIN, their future value is some multiple of their first purchase.

What is the general approach to this? Two separate models with the output of one logistic regression feeding a GLM? Something else? Looking for some expert opinion here on how to handle. Some bayesian hierarchical model?

  • $\begingroup$ This is called a two-part model in econometrics. $\endgroup$ – Dimitriy V. Masterov Jan 22 at 6:47

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.