Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
5058168 | Economics Letters | 2016 | 4 Pages |
Abstract
â¢This paper introduces a random forests-based early warning system (EWS).â¢Our approach is significantly more accurate than other conventional EWSs.â¢There are 730 banks in danger with assets equivalent to about 95.3 million US dollars in total.
This paper introduces a novel random forests-based early warning system for predicting bank failures. We apply this method to the analysis of bank-level financial statements, in order to find patterns that identify banks in danger of failing. The experimental results show that our method outperforms conventional methods in terms of prediction accuracy.
Related Topics
Social Sciences and Humanities
Economics, Econometrics and Finance
Economics and Econometrics
Authors
Katsuyuki Tanaka, Takuji Kinkyo, Shigeyuki Hamori,