Hi experts out there!
I’m trying to use survival analysis for my user churn prediction. In General, I can get % of survival rate (X axis) as time goes by (Y axis) from survival analysis for one of different cohorts..
However, what I want to find out is X,Y,Z value of this statement - “when a user has (less than) X posts within Y time frame the chances of survival (no survival) is Z.”
Or find A,B,Y,Z value of this statement - “when a user has (less than) A posts and B chats (any different type of event) within Y time frame the chances of survival (no survival) is Z.”
I have been reading time dependent covariates survival analysis so far but still stuck in getting X, A, B value. (Of course, the dataset comes with timestamps for all activities.)
Should I jointly use survival analysis and something else? I really appreciate any guidance on this. thank you so much!