Financial Time Series Analysis of SV Model by Hybrid Monte Carlo.pdf

Financial Time Series Analysis of SV Model by Hybrid Monte Carlo 8 Tetsuya Takaishi 0 0 Hiroshima University of Economics, 2 Hiroshima 731-0192 JAPAN l E-mail: takaishi@hiroshima-u.ac.jp u J 8 Abstract. We apply the hybrid Monte Carlo (HMC) algorithm to the 2 financial time sires analysis of the stochastic volatility (SV) model for the ] first time. The HMC algorithm is used for the Markov chain Monte Carlo h (MCMC) update of volatility variables of the SV model in the Bayesian p inference. We compute parameters of the SV model from the artificial c- financial data and compare the results from the HMC algorithm with o those from the Metropolis algorithm. We find that the HMC decorrelates s the volatility variables faster than the Metropolis algorithm. We also s. make an empirical analysis based on the Yen/Dollar exchange rates. c i Key words: Hybrid Monte Carlo Algorithm, Stochastic Volatility Model, s y Markov Chain Monte Carlo, Bayesian Inference, Financial Data Analysis h p [ 1 Introduction 1 v It is well known that financial time series of asset returns shows various inter- 4 esting properties which can not be explained from the assumption of that the 9 time series o