Detecting Jumps Based On High Frequency Financial Data

In the capital market and the financial field,the uncertainty and volatility of the asset price is the root,of the market profit-taking,and it is also the fundamental of the long-term active financial market.Based on the change of the asset price,it has become a hot topic in the financial industry.A large number of research literature on investment and risk management is based on the basic assumptions that the asset price process obeys the continuous path.In fact,there are often some extreme events in the financial market.In addition,it has been found that changes in asset prices are also affected by other factors(such as policy reform,social impact of events,etc),so the researchers suggested that the asset price model in the study to join the consideration of extreme events and other influencing factors,that is called as jumps.Based on the pre-averaging method,this paper constructs a statistical method to detect the jump behavior in the process of asset price,which makes the statistical method more feasible and presents the method of checking the jumps in the opti-mized financial data.The main work and innovation of this paper are summarized as follows:1.Based on the pre-averaging method,construct the test statistic of detecting the existence of the jumps,and discuss the limit property of the test statistic.In this paper,The method of detecting jumps is theoretically established.2.This paper takes into account the noise process in financial high-frequency data,and then we discussed the effect of jumping on the existence of jumps under different noise processes,and makes the results more reliable.At the same time,the last part of the article is based on real stock history data.The simulation results show the effectiveness of this method.