I have 3 different datasets in which customer can or cant be present in all 3 datasets. I have built 3 models for all 3 datasets which are working fine. Kindly note, i cant make a make model combining all dataset together because not all customers would be present in all dataset due to which many variables would be null if any customer is present in the other 2 datasets.

Now, I need to create a single score for each customer by combining output of these 3 models. Output is currently in deciles. At the end, I should be able to use this combined single score to take decisions not individual model scores.

enter image description here


  • $\begingroup$ Welcome to SE! I understand your problem but what is the question you are asking? $\endgroup$ Jul 28 at 4:17
  • $\begingroup$ I need some method or technique that can combine the individual model output into one . If you will see table 2, same customer has different Risk Deciles from different datasets. i need a single metric against each customer. $\endgroup$ Jul 28 at 6:58
  • $\begingroup$ I see. You can average out the output from each model. So C1 = 2, C2 = 5, C3 = 7, and so one. This is common when building ensemble models. $\endgroup$ Jul 28 at 13:29

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.