Default Probabilities: Three different models of estimating defult probabilities

Sammanfattning: The aim of this study is to examine three different ways of estimating the probability of default; one based on historical accounting data, as by Skogsvik 1987, one based on the implied probabilities from credit ratings and one model that estimates the implied probability of default from the pricing of 1 year Credit Default Swaps (CDS). The sample consists of 15 large Swedish companies with traded CDSs over the time period of January 1, 2005 to March 31, 2008. The conclusion reached is that the large increase of the probability of default as suggested by the increase in CDS spreads is not supported by the other models. Several possible explanations for this are identified and discussed. The second conclusion is that the Rating agencies and Probit Model both show much more stable results for the entire period, on significantly lower and more reasonable levels then the CDS Model. Thirdly one could state that as none of the companies in the sample have defaulted the Probit Model offers the best default indication.