A New Standard Of Effectiveness Of Moving Average Technical Indicators In Stock Markets

Considering the real world economic significance of the stock technical indicators, it is proved that the criterion set out in this thesis is different with criterions used in former such as, a technical indicator is effective if the average daily return of holding days or short days under the strategy defined by the technical indicator is significantly bigger than the average daily return of the whole period and so on. The criterion in this thesis emphasizes the importance of investors' time-cost and chance-cost. 'Rational' people will apply technical indicators to investment only when they are effective by this new standard and influence the efficiency of markets as a result. Because of the specialty of financial data, bootstrap methodology is utilized to test the effectiveness of moving average indicators in Nasdaq index during 1993 to 2006.Secondly, the returns of Moving Average technical indicators are simulated by Matlab Software. Then the assumption in the thesis that the effectiveness of technical indicators correlates with the characters of Markets are proved significant in statistics. Moreover, the empirical probability of technical indicators' excess return is proved to change along with the risk structure of the market changes which is represented by the volatility of main stock indexes in the market. The linear correlation coefficient of the indicators' return series, the standard deviation series of the indicators' return and the historical volatility of stock indexes is also studied in this thesis. It is discovered that technical indicators whose linear correlation coefficients defined before are significantly high do exist. Furthermore the signs and the numbers of these coefficients change as the risk structures of Markets change.