A0194
Title: A modified VAR-GARCH model for asynchronous multivariate financial time series via variational Bayesian inference
Authors: Wei-Ting Lai - National Central University (Taiwan) [presenting]
Abstract: A modified VAR-GARCH model, called M-VAR-GARCH, is proposed for modeling asynchronous multivariate financial time series with GARCH effects and simultaneously accommodating the latest market information. A variational Bayesian (VB) procedure is developed to infer the M-VAR-GARCH model for structure selection and parameter estimation. We conduct extensive simulations and empirical studies to evaluate the fitting and forecasting performances of the M-VAR-GARCH model. The simulation results reveal that the proposed VB procedure produces satisfactory selection performances. In addition, our empirical studies find that the latest market information in Asia can provide helpful information to predict market trends in Europe and South Africa, especially when momentous events occur.
