The Strategy Of Dynamic Statistical Arbitrage For China Stock Index Futures Based On High Frequency Data

After20years of development, the financial market has made brilliantachievements. Due to the short time of development, China’s financial market is notmature, and it has great market fluctuations. Therefore, the trading strategy with lowrisk and stable returns has gained great concerns from the business and academic field.Since the listing of stock index futures in2010and the liberalization of the securitiescredit trading, China’s financial market has embraced the era of bilateral trade. Also themechanism has provided the foundation for the realization of statistical arbitrage tradingstrategies.This paper aims to study the statistical arbitrage trading strategies of China’s stockindex futures market. It firstly makes clear definition for some concepts in this article.Then it describes the stock index futures pricing theory from the core of arbitrage. Thetheoretical basis of existing arbitrage opportunities has been analyzed from thepespectives of arbitrage price distortion and value return. After analyzing the domesticand foreign core arbitrage trading models and trading strategies, this paper studies thecalendar spread arbitrage of stock index futures contracts with different maturitiesthrough the co-integration model and statistical arbitrage, and it does reasearches on thespot-futures arbitrage between stock index futures and ETF portfolio with the samemethod.This paper builds calendar spread arbitrage model of stock index futures suitablefor the Chinese market. It carries on the empirical research with one minutehigh-frequency data which are more accurate to the real operation. The validity of themodel and the actual transaction effects have been empirically tested by the dynamicmethods. The empirical results show that the entire trading period, from February28,2013to April2, containing25trading days, there are2110times of arbitrageopportunities,1502of which have been succeeded and608of which have been stoppedloss. The success rate is71.18%. The average holding period is8.5minutes, and thelongest holding time is77minutes. The successful overall yield is44.2%. The arbitragetrading strategies obtain a relatively good income level. This paper builds the stock index futures-spot model of Chinese marketconsidering the actual problem of China’s financial market trading mechanism. Itfocuses on the constructure of ETF portfolio to make it perfectly match for stock indexfutures. It carries on the empirical research with one minute high-frequency data whichare more accurate to the real operation. The validity of the model and the actualtransaction effects have been empirically tested by the dynamic methods. The empiricalresults show that the entire trading period, from August1,2012to August21, containing15trading days, there are1369times of arbitrage opportunities,1080of which havebeen succeeded and289of which have been stopped loss. The success rate is78.8%.The average holding period is64minutes, and the longest holding time is255. Thesuccessful overall yield is23.18%. The arbitrage trading strategies obtain a relativelygood income level.