A0242
Title: Semiparametric Bayesian forecasting for copula stochastic volatility model
Authors: Audrone Virbickaite - CUNEF Universidad (Spain) [presenting]
Martina Danielova Zaharieva - Erasmus University Rotterdam (Netherlands)
Fabian Goessling - University of Muenster (Germany)
Abstract: The contribution is twofold. First, a new highly flexible semiparametric copula stochastic volatility model (SCSV) is proposed, which is based on an infinite location-scale mixture at the financial return level, and accounts for the asymmetric volatility persistence by using a copula-based transition at the latent volatility level. Second, we propose a new and highly flexible Bayesian sampling algorithm for nonlinear state space models under nonparametric distributions. The estimation framework combines a particle filtering and smoothing algorithm for the latent process with a Dirichlet process mixture model for the error term of the observable variables. In particular, we overcome the problem of the intractable likelihood of the newly proposed SCSV model and avoid constraining the model by transformations or the need for conjugate distributions. We test our approach for several nested model specifications using simulated data and provide return density forecasts. Finally, we present an application study using financial returns of several stock market indices.
