แบบพยากรณ์การล้มละลายของสถาบันการเงินในประเทศสหรัฐอเมริกา : ประยุกต์ใช้การวิเคราะห์การถดถอยโลจิสติก (The Prediction Of Financial Institutions Bankruptcy In The U.S.: The Application Of Logistic Regression Analysis)

Abstract

This research aims to study the factors affecting the financial company’s bankruptcy. In this case, the sample selected was divided into 2 groups: 1) bankrupt companies and 2) operating companies. All these companies are financial institutions registered in U.S. and has total assets value more than $1 billion, total 60 companies and during the 2007 – 2009, that was subprime crisis which cause damage to the financial sector in U.S. and this impact spread to financial sector in other countries around the world. The variables in this study included 17 financial ratios and the statistics tools Logistic regression (LOGIT) are used in the study. The results from LOGIT analysis was significantly variables (financial ratios) which can be variables in prediction model are 1) Total equity/total assets 2) Provision for loan losses/gross loans 3) Net income/total assets and 4) Cash/total assets. The model from results correctly classifies 96.67% of the sample at significant 95%. There results can be extended to a general case of financial institutions failure and useful for predict and protect company from bankrupt in the future.