A Study of Panel Logit Model and Adaptive Neuro-Fuzzy Inference System in the Prediction of Financial Distress Periods
University of London, Royal Holloway College - Department of Economics
September 26, 2010
World Academy of Science, Engineering and Technology, Vol. 64, 2010
The purpose of this paper is to present two different approaches of financial distress pre-warning models appropriate for risk supervisors, investors and policy makers. We examine a sample of the financial institutions and electronic companies of Taiwan Security Exchange (TSE) market from 2002 through 2008. We present a binary logistic regression with paned data analysis. With the pooled binary logistic regression we build a model including more variables in the regression than with random effects, while the in-sample and out-sample forecasting performance is higher in random effects estimation than in pooled regression. On the other hand we estimate an Adaptive Neuro-Fuzzy Inference System (ANFIS) with Gaussian and Generalized Bell (Gbell) functions and we find that ANFIS outperforms significant Logit regressions in both in-sample and out-of-sample periods, indicating that ANFIS is a more appropriate tool for financial risk managers and for the economic policy makers in central banks and national statistical services.
Number of Pages in PDF File: 7
Keywords: ANFIS, Binary Logistic Regression, Financial Distress, Panel Data
JEL Classification: C33, C35, C45, C53, C63Accepted Paper Series
Date posted: September 26, 2010 ; Last revised: November 9, 2010
© 2013 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo8 in 0.812 seconds