The Empirical Analysis Of Realized Jump Test And Volatility Model

Past decades,The information technology had archived great development,more faster computer capability,more cheaper cost of storage,Acquirement of trading data is becoming more easier.More and more scholars join into the researching of this topic.This study totally turn into a new hot topic in financial field.High frequency data contain more market information.In this field,volatility has always been the focus of scholars' research.Based on the high frequency data of the Shanghai and Shenzhen 300 index,this paper study the high-frequency data of 3 periods.Before the introduction of stock index futures,after the introduction of stock index futures,the recent,after stock connected between Shenzhen,and Hong Kong.The high-frequency data is sampled in every 5 minutes.The main research contents are as follows:1.There are two important events.The first is the introduction of stock index futures,and the other is stock connection between Hong Kong and Mainland.We study how these change the volatility of The 300 index.We find that the volatility of market is reduced by the launching of stock index future.,but the connection of stocks make it come back.We estimate 4 kinds of volatility measure,RV,RBV,MedRV and MinRV.2.In this paper,we try to improve the HEAVY model with a new realized volatility measure,realized semi variance.And compare this model(HEAVY-RSV)with HEAVY-GJR,they are merely equal.But HEAVY-RSV need less parameters,which could reduce the difficulty of calculating.