Deep learning models for bankruptcy prediction using textual disclosures

Feng Mai; Shaonan Tian; Chihoon Lee; Ling Ma

doi:10.1016/j.ejor.2018.10.024

Back

Deep learning models for bankruptcy prediction using textual disclosures

Journal article

Peer reviewed

Deep learning models for bankruptcy prediction using textual disclosures

Feng Mai, Shaonan Tian, Chihoon Lee and Ling Ma

European journal of operational research, Vol.274(2), pp.743-758

04/16/2019

DOI: 10.1016/j.ejor.2018.10.024

View Online

Abstract

•We propose deep learning models for bankruptcy predictions using textual disclosures.•We compare the average embedding model and the convolutional neural network.•We show that deep learning can combine textual and numerical data for prediction.•We use the erasure method to find the important words for bankruptcy prediction. This study introduces deep learning models for corporate bankruptcy forecasting using textual disclosures. Although textual data are common, it is rarely considered in the financial decision support models. Deep learning uses layers of neural networks to extract features from textual data for prediction. We construct a comprehensive bankruptcy database of 11,827 U.S. public companies and show that deep learning models yield superior prediction performance in forecasting bankruptcy using textual disclosures. When textual data are used in conjunction with traditional accounting-based ratio and market-based variables, deep learning models can further improve the prediction accuracy. We also investigate the effectiveness of two deep learning architectures. Interestingly, our empirical results show that simpler models such as averaging embedding are more effective than convolutional neural networks. Our results provide the first large-sample evidence for the predictive power of textual disclosures.

Bankruptcy prediction

Decision support systems

Deep learning

Machine learning

Textual data

Details

Title: Subtitle: Deep learning models for bankruptcy prediction using textual disclosures
Creators: Feng Mai - Stevens Institute of Technology
Shaonan Tian - San Jose State University
Chihoon Lee - Stevens Institute of Technology
Ling Ma - Stevens Institute of Technology
Resource Type: Journal article
Publication Details: European journal of operational research, Vol.274(2), pp.743-758
Publisher: Elsevier B.V
DOI: 10.1016/j.ejor.2018.10.024
ISSN: 0377-2217
eISSN: 1872-6860
Number of pages: 16
Language: English
Date published: 04/16/2019
Academic Unit: Business Analytics
Record Identifier: 9984701831902771

Metrics

1 Record Views