The Role of Recession Forecasts and F-score in Predicting Credit Risks

Abstract

An accurate forecast for corporate default risk is crucial for financial institutions to effectively quantify potential expected and unexpected borrower losses. This paper develops a logistic regression model to simultaneously examine the roles of firm-specific fundamentals and economic outlook in predicting corporate default risk. The model is estimated using U.S. listed non-financial firms over the 1997⎼2009 period. The empirical findings suggest that economic outlook captured by recession probability forecast, firm-specific fundamental factors measured by the Z-score and F-score, and the interaction between economic outlook and firm-specific fundamental factors play statistically significant roles in predicting bankruptcy risk. The model exhibits relatively high levels of goodness-of-fit and classification ability, and low levels of forecast error.