Market Volatility,Intraday Jump And High-Dimensional Covariance Risk

Research on the price fluctuation risk of asset(s)has been a hot issue in academic research,and the volatility,jump and covariance risk are three important areas in the study of it.With the rapid development of high-frequency trading and program trading,it has prompted scholars to use high-frequency data to study the risk.Nowadays,various “realized” measures such as realized volatility,realized jump,and realized covariance e.t.c.,which can measure different types of risk directly,have sprung up.Therefore,the thesis analyzes four important issues about the price fluctuation risk in the China’s financial market.Specifically,the main contents of this paper are as follows:First,the thesis has studied the effect of market quality on the volatility asymmetry.Particularly,we use a high-frequency-based nose variance estimator as the comprehensive proxy of the market quality and find that volatility asymmetry is closely related to it.Empirical results show that the volatility asymmetry will vanish or even be reversed in the period of poor market quality,which is mainly due to the sharp decline of the leverage effect.Moreover,the volatility feedback effect will be enhanced while the leverage effect will be weakened if the noise variance is taken into consideration in the causal analysis.Finally,we make the robustness analysis and get similar results.Second,the thesis has investigated the effect of different trading behaviors in the stock index futures market on the stability of the stock index futures and spot markets.Particularly,based on the high-frequency data in China’s stock spot and futures markets,we analyze the impact of three major trading behaviors in the futures market,i.e.,hedging,speculation and arbitrage,on the volatility and relation risk of the spot and futures markets.We find the effect of hedging demand and that of speculation strength and arbitrage opportunity are quite different: if the demand of hedging trading increases,it will reduce the volatility of the spot market.However,if the strength of speculation trading or the opportunity of arbitrage trading in the futures market increases,it will exacerbate the volatility of the spot and futures markets to some extent.In addition,we also find that the hedging trading demand has a significant impact on the correlation risk for the spot-futures relationship.Third,the thesis has studied the information spillover effect of the intra-day jump risks between the stock index spot and futures markets.Particularly,we analyze the information spillover effects of jump risk by using the high-frequency data.We find that downward jump risk overflows with each other regardless of the market state,but the spillover from futures to spot is more significant.Morveover,there are significant information spillovers for the upward jump risk from futures to spot,but the effect from spot to futures is affected by the market states.Finally,this thesis has provided a new approach to forecasting high-dimensional covariance matrices based on a principal component analysis of high-frequency data.Our method can avoid the so-called "curse of dimensionality" and handle the case that the number of assets is time-varying.In particular,we propose four(V)HAR-type dynamic models for predicting high-dimensional covariance matrices.All of them can well characterize the long memory behavior of realized eigenvalue series and be easily estimated by OLS.The empirical evidence shows that they outperform the competing models without consideration of long memory behavior in terms of in-sample fitting,out-of-sample prediction,and out-of-sample portfolio allocation.