Estimation of a Stochastic-Volatility Jump-Diffusion Model
Abstract
This paper makes two contributions: (1) it presents estimates of a continuous-time stochastic-volatility jump-diffusion process (SVJD) using a simulation-based estimator, and (2) it shows that misspecified models that allow for jumps, but not stochastic volatility, can give very bad estimates of the true process.Simulation-based estimation is a very flexible and powerful technique. It is ideally suited to high frequency financial data. It can estimate models with intractable likelihood functions, and since the simulations can be performed in (essentially) continuous-time the estimates are consistent estimates of the parameters of the continuous-time process.
Upon submission of an article, authors are asked to indicate their agreement to abide by an open-access license. The license permits any user to download, print out, extract, archive, and distribute the article, so long as appropriate credit is given to the authors of the work. The license ensures that your article will be as widely available as possible and that your article can be included in any scientific archive. Please read about the Creative Commons Attribution License before submitting your paper.
Except where otherwise noted, content on this site is licensed under a Creative Commons Attribution 3.0 License
