AJM - Abstracts

Bankruptcy Prediction: Application of Logit Analysis in Export Credit Risks

Li-Chiu Chi and Tseng-Chung Tang

Abstract

To date, relatively little empirical research has been conducted on the efficacy of the trade credit risk prediction model in the context of international trade applications. Using a sample of listed firms in seven Asia-Pacific capital markets (Hong Kong, Japan, Korea, Malaysia, Singapore, Thailand, and the Philippines) from 2001 to 2003 with available data, we have made a preliminary attempt at empirically studying a predictive export credit risk model based on financial ratios, firm-specific characteristics (size, maturity, R&D expenses, and depreciation expenses), and country risk measures. The results show that our Logit models demonstrate decent classification accuracy and robustness. Specifically, the prediction ability is approximately equal to classification ability when the model is applied to a testing sample. Furthermore, the results indicate that the closer the analysis is to the credit crisis occurrence, the more improved the classification accuracy and prediction accuracy are.

Download this article.

Keywords

BANKRUPTCY PREDICTION; EXPORT CREDIT RISK PREDICTION; LOGIT ANALYSIS.

Contact Details

Li-Chiu Chi and Tseng-Chung Tang
Department of Finance, National Formosa University,
4 Wenhwa Road, Huwei, Yunlin County 632, Taiwan.
E-mail: stella@nfu.edu.tw, tctang@nfu.edu.tw

Helpful comments and suggestions from an anonymous referee are gratefully acknowledged. Partial financial support was received from the National Science Council, Taiwan, ROC, under Projects NSC 89-2626-H-157-001 and NSC 90-2626-H-157-002.