2021 Fiscal Year Final Research Report
Business failures prediction by the machine learning
Project/Area Number |
19K01843
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 07080:Business administration-related
|
Research Institution | Nihon University |
Principal Investigator |
OTSUKI Akira 日本大学, 経済学部, 教授 (30527833)
|
Project Period (FY) |
2019-04-01 – 2022-03-31
|
Keywords | 機械学習 / 企業倒産予測 |
Outline of Final Research Achievements |
This study did developed the model of company bankrupt by machine learning model using chronological order financial statement data. Previous studies are for predicting binary and therefore, they can predict only binary, it will go bankrupt or will not, by machine learning from the concerned financial statement data. However, if the business organization does not understand what is the cause of going bankrupt, it cannot find any measures to take in the first place. Accordingly, in this study research and development was made about the models to clarify the financial statement data which can be the cause of bankruptcy in addition to just the prediction that it will go bankrupt or will not.
|
Free Research Field |
機械学習
|
Academic Significance and Societal Importance of the Research Achievements |
従来の倒産企業予測モデルでは,判別分析やロジスティク回帰分析を用いて財務諸表データを分析するものが多く,つまりは過去の財務諸表データを分析して「倒産する,しない」の2値を予測することしかできなかった.しかし,「何が原因で倒産しそうなのか?」という理由が分からなければ企業は対策の打ちようが無い. ゆえに,本研究では,この2値に加え,倒産の原因になり得る財務状態についても明らかにすることができるモデルを研究開発した.倒産する,しない,の予測だけでなく,倒産の原因になり得る財務状態についても明らかにすることができるようになるため,企業にとって有用な知見が得られるようになると考えられる.
|