Research On Jump Behavior Of A+H Cross-listed Stock Price

The behavior of stock price jump is a focal issue in the Market Microstructure theory,as an important part of the movement of the financial assets prices.Jump phenomenon is common in the stock markets.It has a great impact on the volatility estimation.Therefore,research on jump behavior has important significance on portfolio allocation,asset pricing and risk management.In this paper,we choose AHA index,AHH index and six cross-listed company stocks as research objects and research on jump behavior of financial assets based on realized measurement methods.The main work and conclusions of this paper are listed as follows:Firstly,research on jumps identification and characterization.Based on the L-M jumps test,we use high-frequency financial data to analyze the jump behavior.The results indicate that the A-share market tends to happen big jumps and the H-share market tends to happen small jumps frequently.The results show that the proportion of common jumps is very low and the jumps behavior of two markets appear obvious intraday patten,week mode and asymmetry.Furthermore,jump intensity is a self-exciting process has been verified.Secondly,construct and evaluate the volatility forecasting models.We construct HAR-RV model?HAR-RV-CJ model?LHAR-RV-CJ model and LHAR-RV-CJ-I model and find that volatility of the stock market presents clustering and short-term asymmetry.Based on rolling time windows,we evaluate out-of-sample ability of volatility forecast models by M-Z regression and MCS method.The results indicate although there is not an excellent model,LHAR-RV-CJ-I models containing long memory,leverage effect and jump intensity component is best in summary,especially in the A-share market.Thirdly,research on the relationship between jump behavior and economic information.The results of instantaneous jumps intensity model and Tobit model show that the release of the information makes a difference on jump intensity and information shocks have an asymmetric impact on jump yields.Through horizontal comparison,we can find that economic information has a greater impact on A-share market than H-share market and macroeconomic information has a greater impact on index than individual stock.Finally,research on the relationship between jump behavior and liquidity.We make use of the intraday event study to analysis liquidity dynamic behavior around the jump.Conclusions show that real-time transaction costs and trading volume are not affected by the upcoming jumps,but rose sharply when jump arrives,in other words,the stock markets have depth,but lack of width when jump happens.After jump occurred,the width of the market quickly recover to the lever of no jump,but markets still have depth.Moreover,probit model estimation results indicate that with the real-time transaction costs and trading volume increase,the probability of occurrence of the jump will increase.