Analysis Of The Pairing Trading Strategy Of The Shanghai And Shenzhen 300 Stock Index Stocks

From the end of 2017 to the beginning of 2018,accompanied by the market and policy-oriented changes such as the sharp decline of A-shares and tightened supervision,the market can provide investors with stable and substantial assets,such as non-standard debts.More and more,in response to changes in market style,the steady investment strategy that was not popular in the bull market has returned to people's perspective.In recent years,more and more paired trading strategies have been applied to the A-share market.Paired trading is one of the methods of using relative value investment strategies.In foreign securities markets,it is a widely used statistical arbitrage investment strategy.Paired short and long positions in stocks,earning the convergence of stock spreads.A significant advantage of the paired trading strategy is that the hedging mechanism effectively circumvents the systemic risk of investment,and even in the period when the market is generally down,the paired transaction can still obtain relatively stable returns.The most important of the paired trading strategies is two aspects.The first is how to select the best pairing combinations and the related trading models.The second is how to formulate the optimal trading plans so that the trading utility function can be optimized.The three commonly used methods at home and abroad are cointegration theory,random spread method and minimum distance method.In this paper,the most commonly used cointegration theory and Copula theory,which is often used in the research of financial risk,are selected to compare and analyze the same research target selected by correlation method combined with the least spread square sum method.When using the cointegration theory,although some time series of economic variables are non-stationary,their linear combinations may be stable.If there is a long-term and stable equilibrium relationship among these variables,they are said to have a cointegration relationship.By constructing the residual sequence of spreads,a multiple of the standard deviation of the residual sequence with respect to its standard deviation is calculated to construct a trading signal,and the paired trading operation is discriminated.When Copula theory is used,the ARMA-GARCH model is used to fit the distribution of the stock returns,and after the two stocks are constructed by the Copula function to construct the joint distribution parameters,the best connection function is selected,and the yield sequence is derived by derivation.The conditional probability signal between the two can be used to discriminate the probability that a stock is overvalued or undervalued relative to another stock at a certain point in time,and a cumulative trading signal is constructed by superimposing the conditional probability signal to further perform trade discrimination signals.Dynamic optimization to determine the trend of deviation between two stocks.Finally,by calculating the returns,Sharpe ratio,and maximum retracement of the two strategies,we find that the more complex Copula theory method of trading signal design in this paper has a better performance than the cointegration theory method.Through a comprehensive analysis of the paired trading scheme of this paper,the approach and rationality of the implementation of the program are fully demonstrated,and the conclusion that the paired trading scheme is feasible in the current market is obtained.