When You Can’t Measure What Matters: A Closed-Loop Harness for Probability-of-Default Estimation
Author(s): Hossain Pazooki Originally published on Towards AI. When You Can’t Measure What Matters: A Closed-Loop Harness for Probability-of-Default Estimation Causal estimators for loan decisions cannot be graded on real lending data, because the data never contains the answer. Planting the truth …