'Kaplan Meier survival curve evaluation
I have generated a Kaplan Meier survival curve on the consumer data (the event of interest is 'Churn'). I have the survival curve for both buyers and nonbuyers. Before jumping into the use of the model. I want to know how I can evaluate the truthness of the curve?
I have already tried creating a separate curve for the two different consumers cohorts (who joined in a different year) for the span of 36 months. I noticed that these curves are not similar at all. I believe this is not the right way to evaluate. Can somebody tell me what can be tried to evaluate the survival curve apart from the statistical methods?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|