Particle Filters for Markov Switching Stochastic Volatility Models

Research Paper Number: 299, Quantitative Finance Research Centre, University of Technology, Sydney

21 Pages Posted: 23 Oct 2012

See all articles by Yun Bao

Yun Bao

affiliation not provided to SSRN

Carl Chiarella

University of Technology, Sydney - UTS Business School, Finance Discipline Group

Boda Kang

AMP

Date Written: January 18, 2012

Abstract

This paper proposes an auxiliary particle filter algorithm for inference in regime switching stochastic volatility models in which the regime state is governed by a first-order Markov chain. We proposes an ongoing updated Dirichlet distribution to estimate the transition probabilities of the Markov chain in the auxiliary particle filter. A simulation-based algorithm is presented for the method which demonstrated that we are able to estimate a class of models in which the probability that the system state transits from one regime to a different regime is relatively high. The methodology is implemented to analyze a real time series: the foreign exchange rate of Australian dollars vs South Korean won.

Keywords: particle filters, Markov switching stochastic volatility models, sequential Monte Carlo simulation

JEL Classification: C61, D11

Suggested Citation

Bao, Yun and Chiarella, Carl and Kang, Boda, Particle Filters for Markov Switching Stochastic Volatility Models (January 18, 2012). Research Paper Number: 299, Quantitative Finance Research Centre, University of Technology, Sydney, Available at SSRN: https://ssrn.com/abstract=2163902 or http://dx.doi.org/10.2139/ssrn.2163902

Yun Bao

affiliation not provided to SSRN ( email )

Carl Chiarella (Contact Author)

University of Technology, Sydney - UTS Business School, Finance Discipline Group ( email )

PO Box 123
Broadway, NSW 2007
Australia
+61 2 9514 7719 (Phone)
+61 2 9514 7711 (Fax)

HOME PAGE: http://www.business.uts.edu.au/finance/

Boda Kang

AMP ( email )

Sydney, NSW
Australia
0430976988 (Phone)
2154 (Fax)

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
135
Abstract Views
812
rank
235,132
PlumX Metrics