A new perspective of performance comparison among machine learning algorithms for financial distress prediction

Yu Pei Huang, Meng-Feng Yen

Research output: Contribution to journalArticle

Abstract

We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.

Original languageEnglish
Article number105663
JournalApplied Soft Computing Journal
Volume83
DOIs
Publication statusPublished - 2019 Oct 1

Fingerprint

Bayesian networks
Learning algorithms
Support vector machines
Learning systems
Classifiers
Fuzzy clustering
Data storage equipment

All Science Journal Classification (ASJC) codes

  • Software

Cite this

@article{f9fa4ed014964df78d0e5387229e7a6c,
title = "A new perspective of performance comparison among machine learning algorithms for financial distress prediction",
abstract = "We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.",
author = "Huang, {Yu Pei} and Meng-Feng Yen",
year = "2019",
month = "10",
day = "1",
doi = "10.1016/j.asoc.2019.105663",
language = "English",
volume = "83",
journal = "Applied Soft Computing",
issn = "1568-4946",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - A new perspective of performance comparison among machine learning algorithms for financial distress prediction

AU - Huang, Yu Pei

AU - Yen, Meng-Feng

PY - 2019/10/1

Y1 - 2019/10/1

N2 - We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.

AB - We set out in this study to review a vast amount of recent literature on machine learning (ML) approaches to predicting financial distress (FD), including supervised, unsupervised and hybrid supervised–unsupervised learning algorithms. Four supervised ML models including the traditional support vector machine (SVM), recently developed hybrid associative memory with translation (HACT), hybrid GA-fuzzy clustering and extreme gradient boosting (XGBoost) were compared in prediction performance to the unsupervised classifier deep belief network (DBN) and the hybrid DBN-SVM model, whereby a total of sixteen financial variables were selected from the financial statements of the publicly-listed Taiwanese firms as inputs to the six approaches. Our empirical findings, covering the 2010–2016 sample period, demonstrated that among the four supervised algorithms, the XGBoost provided the most accurate FD prediction. Moreover, the hybrid DBN-SVM model was able to generate more accurate forecasts than the use of either the SVM or the classifier DBN in isolation.

UR - http://www.scopus.com/inward/record.url?scp=85070193904&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070193904&partnerID=8YFLogxK

U2 - 10.1016/j.asoc.2019.105663

DO - 10.1016/j.asoc.2019.105663

M3 - Article

VL - 83

JO - Applied Soft Computing

JF - Applied Soft Computing

SN - 1568-4946

M1 - 105663

ER -