Estimating and pricing correlation of credit deterioration is difficult, but can be handled with standard notions of correlation. The same however is not true for default events. The notion of correlation that one needs to use in dealing with credit default is fundamentally different from the notion of correlation that is useful in dealing with credit deterioration in credit portfolios or instruments. This paper provides a model of credit correlation for credit default events that describes how one can calculate (dynamic) correlations between two series of default events. “Time” is a very important factor, but default data are not measured in equal time intervals at all. Empirical investigation of such data sets needs a new type of model for which we obtain the distribution theory of the implied statistics. The model provided here can also be used as a way to forecast default events in one credit using the default events of other credits.
