Forecasting Corporate Bankruptcy: International Evidence

• Activity Number: 516
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Shaonan Tian* and Yan Yu+
• Companies: University of Cincinnati and University of Cincinnati
• Keywords: Bankruptcy Prediction ; Default Probabilities ; Variable Selection ; Hazard Model ; Logistics Regression ; Global Market

Abstract

Corporate financial distress not only incurs serious financial loss to its creditors but also has a high cost to the society and the country's economy. In this paper, we will focus on the corporate bankruptcy prediction for international market using CompuStat Global database. Firstly, we adopt a robust data mining technique for automatic variable selection for static bankruptcy logistic regression model with cross sectional data. Secondly, with the adaptively chosen covariates from the previous step, we apply dynamic discrete hazard model and investigate its performance for the international market using time varying panel data. Finally, we compare the new model performance among different countries as well as with that from US market data using CompuStat database.