Statistics and Its Interface

Volume 7 (2014)

Number 3

Special Issue on Extreme Theory and Application (Part I)

Guest Editors: Yazhen Wang and Zhengjun Zhang

Pricing synthetic CDO with MGB2 distribution

Pages: 309 – 318



Qiurong Cui (Department of Statistics, University of Wisconsin, Madison, Wisc., U.S.A.)

Yong Ma (College of Finance and Statistics, Hunan University, Changsha, China)


In this paper we apply MGB2 distribution to price synthetic CDO. MGB2 distribution has flexible dependence structure and it is suitable to model extreme risk. The monotonicity of the spread of equity tranche with respect to some parameter is shown. We compare our model with the onefactor Gaussian, Clayton and double $t$ models. Although our proposed MGB2 model is not flexible enough to produce the implied compound correlation smile, it is much more flexible to produce the patterns of base correlation curve than the others. Besides, concerning base correlation, MGB2 and double $t$ model match the market data better than the Gaussian and Clayton copula models.


MGB2 distribution, synthetic CDO, correlation smile, base correlation

2010 Mathematics Subject Classification

Primary 60E05, 91Gxx. Secondary 62G32.

Full Text (PDF format)