Investigating Accounting Patterns for Bankruptcy and Filing Outcome Prediction using Machine Learning Models

Abstract

I study the use of non-linear models and accounting inputs to predict the occurrence of litigated bankruptcies and their associated filing outcomes. The main purpose of this study is to identify the accounting patterns associated with bankruptcies. The filing outcomes include, among others, how long the bankruptcy process will endure, whether the firm will successfully emerge after the bankruptcy period, whether the bankruptcy is tortious, and whether it will involve an asset sale. The study highlights the importance of previously unidentified accounting variables that are useful in predicting bankruptcies and bankruptcy outcomes. The study categorises predictor variables in accounting dimensions to empirically identify the importance of each dimension to the prediction tasks. The high dimensionality of the gradient boosting machine allows us to identify and explain the nonlinear interactions between a wide range of variables.

Keywords: Bankruptcy Prediction, Machine Learning, Corporate Default, Bankruptcy Outcomes

JEL Classification: C32, C38, C45, G33