Research On The Effectiveness Of Moving Average Strategy In China

After more than 30 years of development,our country’s A-share market has become the second largest stock market in the world.By the end of 2021,the total market value of A-shares exceeded 90 trillion yuan for the first time,of which listed companies in all sectors are 4697 in total.The size of the market is so large,and there are so many assets for traders and investors to choose from.How to choose stocks and how to choose the timing of buying and selling has always been an important issue for them.Is there a method that can be used in the historical market to earn excessive returns?According to the definition of an efficient market as an invincible market,has my country’s market reached a weak-form efficient market?With these doubts in mind,this paper studies the technical analysis theory for predicting the trend of stock prices.With the help of the moving average trading theory based on Granville’s rule,on the basis of the traditional golden cross and death cross trading strategies,three improved strategies are constructed.Taking the typical and representative CSI 300 index ETF and GEM index ETF in our country’s A-share market as the target,simulate the real trading environment on the Joinquant quantitative platform(considering transaction costs,transaction quantity integers,available capital restrictions and other issues)back-testing to verify whether a certain strategy combination can achieve excess returns in a longer-term historical market.Finally,through the H-M model of the timing ability evaluation model in the fund performance evaluation model and the FF three-factor model in the multifactorial model,an empirical regression test is carried out on the trading returns of the strategy to analyze the source of excess returns.The back test results show that the strategy with some parameter combinations can defeat the market benchmark strategy of "buy and hold" and obtain excess returns.Specifically,the application effect of the traditional gold crossover and death crossover rules is poor,and there will be some improvement after increasing the selling span value,but the strategy under most parameters still has no excess return;Adding long-term or longer-term moving average judgment can significantly improve the performance of the strategy,but adding too-long-term moving average judgment has poor effect;The strategy effect of increasing the recognition of support and pressure of the moving average is unstable.When taking the CSI 300 Index ETF as the target,some parameter combinations are effective,while when taking GEM Index ETF as the target,all parameter combinations are invalid;The effect of increasing the recognition of the "excessive deviation" between the stock price and the moving average is also unstable.When taking the CSI 300 Index ETF as the target,all parameter combinations are effective,while when taking the GEM Index ETF as the target,all parameter combinations are invalid.After empirical regression using HM-FF three factor model,it is found that the average strategy has significant timing ability on the CSI 300 index ETF,while the average strategy does not have significant timing ability on the of GEM index ETF.