I want to model time to churn with a Cox Proportional Hazard model and use it to predict people likely to churn. The question I have is how valid is to calibrate the model with the same people I want to forecast, is't it going to be biased or over-fitted? Just to be clear the data I have is mainly people and their time to churn and people that has not churn. This last group (the people that hasn't churn) is exactly the one that I want to forecast their hazard and time to churn.


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.