Abstract

This paper employs two kinds of data. One is Credit Default Swap (CDS) market quotations in the United States; the other is the corporate bond yield data matched with related CDS market quotations. We estimate the hazard rates of the one-factor squared root process designed for credit risk model with an unscented Kalman filter on two different sets of data. We conduct principal components analysis of the CDS premiums across different reference entities. Then, we conduct the regression tests of the first principal component on two kinds of hazard rates estimated and subsequently, we extract two liquidity factors by calculating the residuals of each regression equation. Empirical examination indicates that the liquidity risk factors estimated in this paper can be good proxies for liquidity risk. We discover that the liquidity risk factor extracted from CDS market quotations combined with corporate bond yield rates has more goodness of fit than the other factor extracted purely from CDS market quotations. It is demonstrated that the liquidity factor extracted from CDS market quotations combined with corporate bond yield rates is more significantly related to interest rate measures than the one extracted from pure CDS market quotations. The results are still the same even after we add in some macroeconomic variables as control variables. Therefore, we conclude that the liquidity factor extracted from CDS market quotations combined with corporate bond yield rates may be a better alternative.  

Keywords

credit default swapcorporate bond yieldliquidity riskKalman filter