I wonder if anyone can help. I have built a predictive churn model in SAS using logistic regression. The gains chart show that the top 3 deciles predict about 60% of churners, so a good model. However the marketing manager wants to know the level of confidence we can expect in achieving this result if we were to roll out the model for a campaign. The manager only wants to test the model if we are 70% confident that we will get 60% of the churners.
Anyone have any ideas? Not sure if the SAS outputs shows such a thing?