The accuracy of measuring credit risk directly decides on the interest on credit, which has to be paid when raising a credit, and the amount of capital to keep in reserve by a firm. The structural credit risk model proposed by Merton (1974) lays the groundwork for the assessment of a firm's credit risk by its default probability. Doubtlessly, the volatility of the firm's equity represents the most sensitive parameter influencing the default probability. By combining the Merton approach with conditional volatility models, we empirically examine in this article that the specification of conditional volatility affects the probability of default and therefor the credit rating. More precisely, we show on German stock market data that financial market data properties (i.e. asymmetric response of conditional volatility to return shocks and long-range dependencies within the conditional volatility) may not be neglected within the computation of credit risk. Moreover, the influence on the default probability by the type of conditional distribution is pointed out.