Java Quant

            Financial Quantitative Algorithms

Jump-Diffusion Bates Model Algorithms

    The Bates Model is a type of Jump-Difussion model that is able to improve calibration results for short term options. The Bates Model consists of Jumps processes built on top a Heston model. For more information visit the link: www.javaquant.net/papers/stochjumpvols.pdf

 

Bates Model: a QuantLib project

The files below are part of a project that I was developing in order to Price Cliquet options using the Log-Jump variant of the Bates model with stochastic volatility (Log-Jump built on top of the Heston Model). I have used QL 0.9.0 (QuantLib) together with C++ Visual Studio 2005 (Express) and Boost 1.3.5. I attached the set of files needed to build the Monte Carlo engine, as well a set of test files where I was checking out the validity of QL code against known result. I'm attaching also a power point presentation and a "toy" implementation of a Cliquet priced with the Bates model on Excel (constant parameters)

Bates Model Analytic Pricer for Vanillas

2/23/2009 - Since we are living in a defaulty world, it came to my interest analyzing if the Bates model calibrated to a set of vanillas options (to GM options for example) would give an average Jump size related to the CDS spread market price. That is: if after calibrating the Bates Model, the average Jump size is proportional to the market quote for CDS spread. The calibration implementation is not imlemented yet. (This model might no suitable for the purpose I'm looking for)

 

MC Cliquet C++ Source Code QL 0.9.0, Boost 1.35, MSVS2005 (express) Download
Test launcher file QL 0.9.0, Boost 1.35, MSVS2005 (express) Download
Several tests on MC Cliquet Bates QL QL 0.9.0, Boost 1.35, MSVS2005 (express) Download
"Toy" Bates Excel 2003 SP3 Download
PPT Presentation: Bates Model Analysis Power Point 2003 SP3 View
Analytical Pricer for Vanillas Java Download

 


www.javaquant.net © 2006 H. Aliaga. Design by Cesar.
XHTML 1.0 Strict