Biologically Inspired Algorithms for Financial Modelling (Natural Computing Series)
Biologically Inspired Algorithms for Financial Modelling (Natural Computing Series)
Calibration of stochastic volatility models on a multi-core CPU cluster
WHPCF '13 Proceedings of the 6th Workshop on High Performance Computational Finance
| Hi-index | 0.00 |
For the analysis of many exotic financial derivatives, the Heston model, a stochastic volatility model, is widely used. Its specific parameters have to be identified from sets of options market data with different strike prices and maturities, leading to a minimization problem for the least square error between the model prices and the market prices. It is intrinsic to the Heston model that this error functional typically exhibits a large number of local minima, therefore techniques from global optimization have to be applied or combined with local optimization techniques to deliver a trustworthy optimum. To achieve results in reasonable time, we approach as follows: (1) For the evaluation of the objective function, we use a Fourier cosine method, optimized for parallelization, and (2) the local/global optimization scheme is carried out on parallel architectures. Results are reported for a multi GPU server and a multicore SGI Altix 4700.