Bayesian Analysis of ARCH-M model with a dynamic latent variable

Abstract

A time-varying coefficient ARCH-in-mean (ARCH-M) model with a dynamic latent variable that follows an AR process is considered. The joint model extends the existing ARCH-M model by considering a dynamic structure of latent variable for examining a latent effect on the time-varying risk–return relationship. A Bayesian approach coped with Markov Chain Monte Carlo algorithm is developed to perform the joint estimation of model parameters and the latent variable. Simulation results show that the proposed inference procedure performs satisfactorily. An application of the proposed method to a financial study of the Chinese stock market is presented.