A Statistical Model for Credit Scoring

摘要: We derive a model for consumer loan default and credit card expenditure.The is based on statistical models discrete choice, incontrast to the usual procedure of linear discriminant analysis. Themodel then extended incorporate probability in modelof expected profit. The technique applied large sample ofapplications expenditure from major company. Thenature data mandates use selection forestimation. empirical profit produces an optimalacceptance rate applications which far higher than theobserved used by vendor discriminantanalysis. I am grateful Terry Seaks valuable comments anearlier draft this paper Jingbin Cao his able researchassistance. provider support project hasrequested anonymity, so must thank them as such. Their help andsupport are gratefully acknowledged. Participants appliedeconometrics workshop at New York University also provided useful commentary.