Distressed Mortgages: A Machine Learning Assessment

Abstract

In this project, we investigate mortgages under distress. Using bank-level loan data from Ireland, we describe mortgage holders that are most likely to engage via a Standard Financial Statement (SFS) as part of the Mortgage Arrears Resolution Process (MARP). For this, we use state-of-the-art machine learning tools. Finally, we tune a machine learning model to predict the probability of distress for a given loan.

Presented at:
- Seminar Day, Central Bank of Ireland (2022)
- Economics Research Seminar, Central Bank of Ireland (2023)

Elio Bolliger

Head of Federal Finances Analysis and Forecasting Team in Financial Statistics

My research interests include international macroeconomics, monetary policy and financial stability.