Bankruptcy Forecasting Review: Focus on South Korea and the U.S.A

Purpose The purpose of this study is to review the bankruptcy prediction related academic articles focusing on both the bibliometrics and survey of selected articles for South Korea and USA. Methods We apply Rstudio’s biblioshiny and VOSviewer to data collected from Scopus and Web of Science databases from 2010 to 2022. Result Our review showed a yearly growth of 4.04% in bankruptcy related literature with the highest articles published in 2019. USA and China have the highest collaboration (9 times) over the sample period. Additionally, bankruptcy prediction models such as logit, hazard models, neural networks and support vector machines (SVMs) have dominated the bankruptcy prediction literature before 2016. Random forest on the contrary has gotten attention from 2018. Conclusion Relatively, a high number of research is done on bankruptcy prediction in South Korea and the USA. Future research on bankruptcy prediction should explore emerging models such as random forest model which is under explored in both South Korea and USA bankruptcy prediction literature. This research contributed to the extant literature by shedding more light on the present status of bankruptcy prediction models for an emerging and a developed economy, which will be useful to academics and other stakeholders.