Defaults are only observed for approved borrowers. How do you value LendScore?
Causal inferenceHardProblem. A lender approves 60% of applicants using a bureau score. They want to know what would happen if they approved the same 60% using LendScore instead. But outcomes exist only for people they approved. How do you estimate the change in defaults?
Before you reveal: say your answer out loud, as if you were in the real interview — get your reasoning across clearly first. There is no single correct answer: reading what the interviewer is really after and defending your own thinking is what makes an answer strong.