The introduction of margin trading and stock index futures provides a good platform for the implementation of statistical arbitrage strategy of Chinese security market. Statistical arbitrage strategy is an important tool for applying the statistical methods to the stock market, high frequency data based statistical arbitrage strategies are able to meet institutional investors arbitrage demand, in the meantime a new and more effective investment is introduced. The statistical arbitrage strategies commonly used are improved and the new statistical arbitrage strategy is designed and tested empirically. The number of arbitrage transactions using the data of each frequency is significantly reduced, the average yield for each transaction is doubled; the purpose for accessing the largest single revenue and reducing transaction frequency is achieved. With new statistical arbitrage strategy, the use of the six frequency data arbitrage is able to achieve very good performance, and the profit forecast of outside samples based on the profitability within the sample is significantly enhanced. With the high frequency data, statistical arbitrage strategy applied to Chinese stock market is effect, and new statistical arbitrage strategy is significantly superior to the common strategy.