Probability bucketing for correlation expansion in CDO pricing

Abstract

Our work is to combine two techniques which are Correlation Expansion technique and Probability Bucketing technique. Correlation Expansion technique shows that CDOs, whose obligors are dependent, can be expressed as a series of prices in independent obligor models. So, this is much less complicated to price CDOs and credit basket derivatives. Probability Bucketing technique is the technique that can create loss distributions. Our purpose is to develop a method of approximating CDOs tranche price and credit basket derivatives, which is less complicated in the computation of the model’s output and improves the calculation speed. We also focus on the way to contribute loss distributions which is one of the important parts for pricing credit derivatives. We find out that this combined method gives us acceptable outputs.