Русский
!

Conference publications

Abstracts

XX conference

The Mathematical Forecasting Methods of The Credit Default Swaps and Copula Models

Stikhova O.V.

Moscow State Technological University “STANKIN”, chair “ The Applied Mathematics Vadkovsky lane 1a, Moscow, 127055, Russia, +7(499)972-95-20, E-mail: olgitast@smtp.ru

1 pp. (accepted)

The credit default swaps (CDS) forecasting and the single and multiple name obligation default probability projection are the actual problem recently. The credit derivatives such the collateralized debt obligation (CDO) are investigated in this work and also we considered the classical default models of the portfolio credit derivatives [1]. There are two parties in the CDS deal: the protection seller and the protection buyer. To quote the base portfolio CDS and CDO tranches our single and multiple name default probability developed models are used respectively. The default time distribution was calibrated according to the basis CDS rates [2]. The CDO and CDS credit derivatives cost definition problems for this paper numerical experiments are considered. Multivariate default distributions are taken into account in models on a copula basis, becoming by the market standard for an estimation of a basket credit derivatives, in particular, on the contracts CDO. We analyzed the portfolio credit derivatives loss distribution models such as the one factor Gaussian copula model, the double normal inverse one factor Gaussian copula model, the same models with the stochastic factors and we considered the large portfolio approximations in these models [3]. With the help of Gaussian copula models from appropriate market data it can be predicted transhes correlation.

A series of computing experiments on market products price parameters modeling of the industrial companies default credit derivatives both with generated samples and real data and the results verification were made. The calculations findings for the industrial producer and also for the various activity field enterprises have shown the developed models and algorithms high efficiency.

References

1. Stikhova O.V. Credit Derivatives Mathematical Modelling. /XLII The National Conference of the Mathematics, Informatics and Chemistry problems. Working Paper. -М: RUDN, 2009, pp.123-124.

1. Vasicek O. The Distribution of Loan Portfolio Value. Risk, 12 (2002).

2. Schönbucher P. J., Schubert D. Copula Dependent Default Risk in Intensity Models.// Working Paper, Department of Statistics, Bonn University, 2001.



© 2004 Designed by Lyceum of Informational Technologies №1533