Article

Numerical Integration Filters for Maximum Likelihood Estimation of Asymmetric Stochastic Volatility Models

Society for Computational Economics, Computing in Economics and Finance 2005 01/2005;
Source: RePEc

ABSTRACT I consider two filtering algorithms (quadrature and mixture Gaussian) based on numerical integration for maximum likelihood estimation of stochastic volatility models with leverage. These algorithms extend straightforwardly to stochastic volatility models with non-Gaussian innovations. A small Monte Carlo simulation experiment shows that the mixture Gaussian filter performs remarkably well both in terms of accuracy and computation time. As an empirical application, I fit the asymmetric stochastic volatility model to the S&P 500 index daily returns with a Gaussian and skew-t innovation. The estimates from the two filtering algorithms are remarkably similar, suggesting the usefulness of the mixture Gaussian filter for practical use

0 0
 · 
0 Bookmarks
 · 
38 Views
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.

Keywords

asymmetric stochastic volatility model
 
empirical application
 
maximum likelihood estimation
 
mixture Gaussian
 
mixture Gaussian filter
 
non-Gaussian innovations
 
numerical integration
 
skew-t innovation
 
small Monte Carlo simulation experiment
 
stochastic volatility models
 
straightforwardly
 
usefulness
 

Hiroyuki Kawakatsu