Research On Option Pricing Of Shanghai Stock Exchange 50ETF Based On Realized Range Double Power Variation

The proposed BSM model implies the establishment of an option pricing theory system,but it has some shortcomings,especially the constant assumption of volatility.Later scholars did a lot of in-depth research on the shortcomings of the BSM model.Improved scholars based on time-varying volatility have successively proposed ARCH and GARCH models,and some scholars have proposed different random volatility model based on the idea of random volatility.In terms of time-varying volatility,early research can only be limited to low-frequency data due to technical problems.Using daily data for research,although it has achieved certain results,it also loses some information about intra-day data fluctuations.In general,Still needs improvement.Nowadays,with the continuous development of science and technology,the difficulty of obtaining high-frequency data within the day has been reduced.Based on a large amount of information about intra-day data fluctuations,scholars have constructed a measure to measure the volatility,which is called the "realized measure".From the first realization of the Realized Volatility,scholars have continuously researched and improved the measurement of high-frequency volatility,and gradually developed some new realized measures.According to relevant theoretical proofs and practical applications,the realized measurement of volatility has effectively used a lot of information contained in high-frequency data,which has excellent properties and practicality.Based on the Realized Volatility,scholars have also proposed RealizedGARCH and other high-frequency volatility models.The Shanghai Stock Exchange officially listed China's first on-the-spot option on February 9,2015,the Shanghai Stock Exchange 50 ETF option.Until December 23,2019,the CSI 300 ETF option was the only domestic on-the-market option.Therefore,in recent years,many domestic scholars have researched on the Shanghai 50 ETF option.In terms of pricing,some scholars have used the traditional B-S-M model,GARCH model,and Heston model to study their pricing issues and achieved certain results.In recent years,some scholars have begun to apply high-frequency volatility models to related research.Common models include Realized-GARCH models and HAR models.Among them,Realized-GARCH has evolved from GARCH models,and HAR models are based on heterogeneity.The core of market theory is a combination of realized measures of volatility.In these studies,scholars mainly use the traditional realized measure of Realized Volatility for modeling analysis.Compared with the Realized Volatility,there are some new realized measures,such as the Realized Range-based Bipower Variation,which theoretically utilizes more information in high-frequency data,so this paper introduces this measure as an improvement to get better results in pricing experiments.In this paper,the above 50 ETF option is used as the research object.Using the high-frequency data within 5 minutes,the Realized Range-based Bipower Variation sequence is calculated.It is introduced into the Realized-GARCH model and the HAR model,and is combined with Monte Carlo simulation method to establish The corresponding option pricing model analyzes the price of the European call options that expire on June 26,2019 with the 50 ETF as the underlying asset.The time span is February 21,2019 to June 26,2019.The results are compared with Realized-GARCH and HAR pricing models based on the revised Realized Volatility.In addition,the traditional B-S-M is used as the basic model to comprehensively analyze the effects of each model.In the pricing empirical study,on the one hand,this paper uses the traditional loss function to judge the model effect,on the other hand,it introduces the SPA advanced prediction ability test method,improves the design according to the research questions,uses more comprehensive option contract data,and splits short-term Test the model pricing effect in the medium and long term.The empirical results show that the highfrequency volatility model has great advantages over BSM;Compared with the pricing model based on Realized Volatility,the accuracy of the pricing model based on the Realized Range-based Bipower Variation is improved,and the medium and long-term forecast performance is more obvious,which shows that the Realized Range-based Bipower Variation on the applicability of the SSE 50 ETF options market is better and is an effective improvement;compared to the Realized-GARCH model,the pricing effect of the HAR model also has certain advantages.This shows that the SSE 50 ETF options market is somewhat heterogeneous and has the influence of different types of investors.In general,the content of this article provides new ideas and improvement points for the research on the Shanghai 50 ETF option pricing,and also puts forward some suggestions and prospects for related research in the future.