Mortgage Delinquency Migration: An Application of Maximum Entropy Econometrics

Defaulting on a mortgage represents the ultimate consequence of past decisions to delay payment. As such, a better understanding of the probability of default is obtainable from better understanding the probability of delinquency that is induced by the sequence of decisions to delay payment. Even so, bank-specific, account-level mortgage data are required to analyze delinquency using conventional statistical methods and these data are rarely available to researchers. This paper introduces a maximum entropy econometric approach that uses publicly available aggregated data to estimate both the probability of delinquency and the probability of default. The results suggest the approach has merit for monitoring bank performance, as well as usefulness for bank's risk management efforts and Basel Accord compliance.