Financial distress forecasting with a machine learning approach

Abstract

A highlighted issue relating to the financial distress of public companies raises more debate from both academic and current practice perspectives as financial markets are currently a key source of growth for the local and international economies. In the context of advanced technology and the digital revolution, forecasting and early detection of financial distress are important methods that contribute to increasing confidence between investors and the market and help to make sound decisions promptly to avoid reaching bankruptcy (Fuentes et al., 2023). This study employs machine learning algorithms to measure the probability of financial distress of listed firms on the Vietnam Stock Exchange by using a dataset with 4,936 observations from 2009 to 2020. The research has identified internal determinants such as debt-to-equity ratio, asset turnover ratio, and profit margin ratio as indicators that have the greatest impact on financial distress under different models. The results reveal that Model 1 — Altman and Model 3 — Zmijewski predict financial distress with an accuracy rate of 98%. In addition, we have determined the threshold when using the decision tree algorithm, which has an important impact on the financial distress of listed firms. This finding contributes to the existing literature review and is consistent with previous studies of Chen et al. (2021) and Martono and Ohwada (2023).

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 16 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.