Bayesian estimation for the threshold stochastic volatility model with generalized hyperbolic skew Student's t distribution

In this article, we study a threshold leverage stochastic volatility model with a generalized hyperbolic skew Student's t distribution (TLSV-GHST). The model can simultaneously capture the leverage effect, skewness, and heavy-tailedness of financial return data. A popular Bayesian method combining the Metropolis-Hastings (MH) algorithm and the Gibbs sampler is developed for the parameter estimation of the TLSV-GHST model. The deviance information criterion is used to assess the fitness of the proposed model. Additionally, we investigate the sensitivity of Bayesian estimates to the priors of parameters. Some simulation studies and examples are presented to illustrate the effectiveness of the proposed method.