Application of Machine Learning in Predicting Financial Distress Among Indonesian Banks

Banks play a crucial role in the economy. Consequently, systemic banking crises destabilize financial markets and hinder global economic growth. In this study, machine learning was used to predict bank distress in the Indonesian banking sector. Key variables relevant to the banking sector were identified. The data, spanning the period 2019 to 2023, was collected from Indonesian banks listed on the Indonesia Stock Exchange (IDX). Random Forest (RF) and XGBoost (XGB) models were employed to develop predictive frameworks, with their performance evaluated using overall prediction accuracy (OPA), Type I error, and Type II error. The results showed that the RF model outperformed the XGB model, achieving the highest accuracy and completely eliminating Type II errors. This study highlights the potential of machine learning to improve early warning systems for financial distress, contributing to the stability of the banking sector and the resilience of the economy.