Design Of Cross-variety Arbitrage Scheme For Stock Index Futures Based On Copula Model

On April 16,2015,China Financial Futures Exchange launched the SSE 50 and CSI 500 stock index futures,plus the CSI 300 listed in 2010.There are currently three major stock index futures,which has enriched the number of existing stock index futures in China.The function of stock index futures has also been added.Prior to this,the statistical arbitrage of stock index futures often only included inter-period arbitrage and futures arbitrage,and the introduction of new stock index futures made cross-species arbitrage possible.Against this background,this article intends to carry out cross-species arbitrage on the three major stock index futures.Based on a systematic study of relevant literature at home and abroad,it is found that scholars at home and abroad mostly arbitrage based on co-integration models.However,because the co-integration model characterizes the linear correlation between the two variables,the position opening signal has certain limitations,which will cause investors to miss some trading opportunities.The main feature of Copula is that it can describe the non-linear relationship between variables,which can increase the chance of capturing transactions.Therefore,this article selects the price sequence of the main index futures contract,performs data processing and connection to form continuous price data,uses statistical and economic principles to analyze,and selects the CSI 300 and SSE 50 to build a statistical arbitrage strategy based on the Copula model.An empirical analysis of cross-varietal arbitrage is carried out,and a common co-integration model is used to compare to verify the feasibility of designing a cross-varietal arbitrage scheme for the Copula model.This article first analyzes the correlation between the processed three major stock index futures price series,that is,the Pearson linear correlation coefficient and the Kendall rank coefficient are obtained for the logarithmic price series.The CSI 300 and Shanghai 50 stock index futures are selected.As a combination of arbitrage trading across varieties;secondly,design trading schemes based on co-integration theory and Copula theory.For co-integration models,calculate their regression coefficients.For Copula models,first use the GARCH model to fit the edge distribution,and then estimate the parameters.Model the model and find itsconditional probability.Finally,design a trading strategy for stock index futures cross-species arbitrage according to the trading rules.Then use the data inside and outside the sample for empirical calculations,calculate the cumulative return rate,maximum drawdown,win rate and other indicators to get arbitrage.Trading results.After empirical findings,the trading signals designed by the Copula model can better capture the trading conditions.Compared with the cointegration model,there are more accurate opening conditions.The data is segmented back-tested,and the cumulative return of the Copula model is Both inside and outside the sample are better than the co-integration model,and the winning ratio of this strategy is more than 60%.The statistical method is used to analyze the cross-species arbitrage of stock index futures,and a more reasonable reference plan is designed for investors.It also has the function of activating China's stock index futures market and promoting the development of the financial futures market.In view of the futures trading itself with high leverage and the statistical arbitrage of only two main stock index futures contracts in this plan,there are certain risks in the design of this plan,mainly:(1)The Copula model may misjudge when predicting the trend of volatility loss.(2)Misjudgment losses caused by impact costs.(3)Incidental shocks,such as changes in the transaction costs of CICC.