Analysis of Volatility Characteristics of Chinese and American Financial Markets Based on SV Model

The stochastic volatility(SV) model is an essential model in the field of financial time series research, which can better characterize the time-varying characteristics of volatility. Based on the Bayesian method of Markov Monte Carlo (MCMC) simulation, we apply stochastic volatility SV-N and SV-T models to conduct empirical research on the volatility of daily return data of Chinese and American stock markets from 2016 to 2021, then evaluate the model through DIC criteria. The empirical results show that, within a given sample period, the yield series of the Chinese and American stock indexes have the characteristics of “spikes and thick tails”, and the volatility level of the US stock market mean is greater than the volatility level of the Chinese stock market, which makes transaction risk higher.