Wrong-Way Risk Correlation Coefficient Calibration

Abstract

Under the new Basel III banking regulation banks should include wrong-way risk (WWR) into the calculation of the credit valuation adjustment (CVA) of the OTC derivatives. WWR takes place when the exposure to a counterparty is adversely correlated with the credit quality of that counterparty. Assuming a link between the interest rate swap (IRS), i.e. financial derivative in which two counterparties repeatedly exchange cash flows based on interest rate value and that the default time is represented by a Gaussian copula with a constant correlation coefficient, the WWR is expressed by this correlation coefficient. Because the observation of the default time means bankruptcy of the company, the correlation cannot be simply estimated using the observed data in contrast to the credit default swap (CDS) rate which is related to the intensity of default. Based on available daily Czech Republic government IRS and CDS rates we estimated the correlation using maximum likelihood method assuming that the systematic factor is governed by the AR(1) process, so we can decorrelate (eliminate autocorrelation) both time series. The results show that the correlation calibrated on the daily data is relatively high, and therefore the WWR should not be neglected in this case.