| Summary: | 碩士 === 元智大學 === 會計學系 === 93 === Abstract
Estimation of asset correlation in the Basel Ⅱ IRB approach have caused much controversy in recent years. Among current available models, however, few adjust asset correlation to reflect cyclical effects. In addition, empirical studies on correlations between borrowers and cyclical patterns in default rates are rather scarce.
In this paper, we use the extended factor model with observable risk factors to examine how macroeconomic risk effects are incorporated into measures of credit risk exposure. Proxies for the credit cycles are provided by U.S. Bureau of Census and Bureau of Labor Statistics. Default rates for each rating are provided by Moody’s investment service. We find that a macro factor model failed to perform well to explicate whether cyclical factors can affect asset correlation or not, suggesting the difficulties in linking macro factors to asset correlations. Finally, we discuss some restrictions and implications for estimation of asset correlations upon credit cycles.
|
|---|
