Research On Pairs Trading Strategy For CSI 300 Constituent Stocks

With the development of computer technology and the prosperity of big data,pairs trading strategy,as a branch of quantitative investment method,has attracted the attention of investors.Pairs trading has a development history of decades abroad and is relatively mature.However,there is a lack of corresponding short selling mechanism in China.It did not begin to develop in China until the emergence of margin trading in2010.This thesis constructs a dynamic trading strategy based on Hurst index,and studies the feasibility and effectiveness of this strategy in the domestic stock market.The model mainly analyzes and studies from two aspects: Stock selection and trading strategy.When selecting stock pairs,firstly,carry out correlation analysis,and sort according to the correlation coefficient from large to small to select the top 100 stock pairs with high correlation;Secondly,ADF test and cointegration test are carried out on the selected100 stock pairs to screen out the stock pairs with cointegration;Then,the cointegration stocks are divided into industries according to the industry division standard of the CSRC;Finally,the Hurst Index is used as an index to rank the stock pairs,and the top three stock pairs from the same industry and different industries are selected respectively.When formulating the trading strategy,firstly,an error correction model is established to calculate the hedging proportion of paired stocks;Then,take 10 trading days as the window,scroll to calculate the time-varying mean and standard deviation,and then get the opening and closing position and stop loss threshold.This model is applied to the data set of CSI 300 constituent stocks.From the return index,the selected stocks obtained 28.25% and 5.43% excess returns respectively,which verified the effectiveness of pairs trading in China's market and the possibility of successful matching of stocks from different industries.From the perspective of risk measurement indicators,the Sharp Ratio of stocks from the same industry is higher than those from different industries,and the Algorithm Volatility and Maximum Drawdown are lower than those from different industries,so the risk is less.