Sixty years of bankruptcy models: issues, limits, and progress

Abstract

During the past 60 years, corporate bankruptcy prediction has become a major
corporate finance research domain. Many papers have been published in various
fields dealing with bankruptcy prediction factors, including modeling, sampling,
data selection, and model performance. However, a clear overview and discussion
of the topic is lacking; this paper fills that gap. Noting the many advances in bankruptcy prediction models in recent years, the authors analyze a large number of
empirical studies, comparing the classification results and/or prediction abilities
of failure prediction models according to various techniques. Two main conclusions
emerge: There is no method that always outperforms all other methods, and the
most important component of bankruptcy prediction models is the selection and
quality of data.