An Empirical Study On The Weak-Form Efficiency Of Shanghai Stock Market

Since the establishment in1990and through22years of exploration,reform and innovation, the stock market in China has made remarkable progress and great achievement. By the end of2011, there are2342listed companies, more than130million investors,106security companies in China’s stock market where the total number of share capital has reached3609.552billion and the total market capitalization has reached2.147581trillion ranking the second in the world.In the past22years, the stock market has played an irreplaceable and very important role in broadening the financing channels of various enterprises, facilitating the transformation of national savings into investment and the capital formation, optimizing the allocation of resources and dispersing market risk, thus effectively boosting the healthy and sustainable development of the real economy in China.With the rapid development of the stock market and the continuous rising of its status in China’s economic and social development, there emerges an increasingly important question that has attracted more and more attention of many parties concerned, including stock investors,listed companies, regulatory departments and finance researchers:To what extent has the efficiency or effectiveness of the stock market in China improved over the history of development?In order to explore the important question, this thesis conducts an empirical study on the weak-form efficiency of Shanghai Stock Market,based on the analysis of predictive ability and profitability of moving average trading strategy. The main content,structure and conclusion of the thesis is summarized as follows below:Firstly,by using MATLAB programming and automatic operation,we investigate the predictive ability and profitability of various moving average trading strategies with different parameters on Shanghai Composite Index from May22th,1992to January31th,2012and find that the moving average trading strategies with the highest average rate of return in the sample period are variable-length moving average trading strategies VMA(S,30,0), S=1,2,3,4,5. We then use a comprehensive set of technical methods and indicators including mean value,t-statistic and Sharpe Ratio to further analyse the pridictive ability and profitability of variable-length moving average trading strategies VMA (S,30,0), S=1,2,3,4,5and find these strategies do display significant and superior pridictive ability and profitability compared with buy-and-hold strategy.Secondly,in order to analyse whether the characteristics of rate of stock market returns which deviate from normal distribution affect the statistical test conducted previously and mitigate the possibly existent data snooping bias problem in the sample data, we conduct bootstrap test to further investigate the pridictive ability and profitability of variable-length moving average trading strategies VMA (S,30,0), S=1,2,3,4,5under the alternative assumptions of return generating process, including Autoregression model AR(n), Generalized AutoRegressive Conditional Heteroskedasticity model AR (n)-GARCH(p,q) and Exponent Generalized AutoRegressive Conditional Heteroskedas-ticity model AR(n)-EGARCH(p,q).We find that the significant and supurior predictive ability and profitability displayed by the variable-length moving average trading strategies VMA (S,30,0), S=1,2,3,4,5can not be explained by the non-normal characteristics of stock market returns including auto regressiveness,time-varying mean value and variance.In other words,after considering the data snooping bias problem in the sample data and the non-normal characteristics of stock market returns, variable-length moving average trading strategies VMA (S,30,0), S=1,2,3,4,5still display significant and supurior pridictive ability and profitability compared with buy-and-hold strategy.Thirdly, we analyse the effect of transaction costs existent in the stock market on the net rate of return of moving average trading strategies.We conduct investor type sensitivity analysis and break-even transaction cost analysis,both concluding that after taking transaction costs into consideration, moving average trading strategies still have considerable potential to make profit and earn abnormal return.Based on the empirical study and Weak-Form Efficient Market Hypothesis,this thesis comes to the conclusion:Shanghai Stock Market has not attained weak form efficiency.Finally, some implication and suggestion is presented as to how to effectively improve the information efficiency of the stock market in China.