A threshold stochastic volatility model

Mike K. P. SO, Wai Keung LI, K. LAM

Research output: Contribution to journalArticles

63 Citations (Scopus)

Abstract

This article introduces a new model to capture simultaneously the mean and variance asymmetries in time series. Threshold non‐linearity is incorporated into the mean and variance specifications of a stochastic volatility model. Bayesian methods are adopted for parameter estimation. Forecasts of volatility and Value‐at‐Risk can also be obtained by sampling from suitable predictive distributions. Simulations demonstrate that the apparent variance asymmetry documented in the literature can be due to the neglect of mean asymmetry. Strong evidence of the mean and variance asymmetries was detected in US and Hong Kong data. Asymmetry in the variance persistence was also discovered in the Hong Kong stock market. Copyright © 2002 John Wiley & Sons, Ltd.
Original languageEnglish
Pages (from-to)473-500
JournalJournal of Forecasting
Volume21
Issue number7
DOIs
Publication statusPublished - Nov 2002

Citation

So, M. K. P., Li, W. K., & Lam, K. (2002). A threshold stochastic volatility model. Journal of Forecasting, 21(7), 473-500. doi: 10.1002/for.840

Keywords

  • ARCH model
  • Gibbs sampling
  • Kalman filter
  • Monte Carlo Markov Chain
  • State space model

Fingerprint Dive into the research topics of 'A threshold stochastic volatility model'. Together they form a unique fingerprint.