Technical Analysis And Asset Pricing

The predictability of the stock market has always been the core issue in financial field.The classic asset pricing theory holds that the market is completely effective,any prediction of the future returns of stocks is in vain.Technical analysis as a way to predict the future trend of the market is not received by the academic field,but in financial industry,technical analysis is very popular.Scholars have done a lot of empirical tests on the effectiveness of technical analysis,most of the empirical results show that technical analysis is effective.Then,behavioral financiers try to explain the effectiveness of technical analysis from a behavioral bias view.Compared with the stock market in developed countries,Chinese stock market is not mature,the system and supervision need to be improved.In addition,Chinese stock market exists many individual investors,their irrational behavior of speculation is outstanding,which reduce market effectiveness.Then,in the Chinese stock market,whether the technical indicators are effective or not,this paper makes an empirical study,and tests differences in the effectiveness of technical analysis under different deviations of behavior.Finally,a new asset pricing model is constructed on the basis of technical indicators to test explanatory power of the stocks' cross section return in A-share market.The main contents and conclusions of this paper are as follows: First,with the frequency trading data of Shanghai Composite Index and Shenzhen composite index during 1997 to 2016,we use the linear regression model in this paper to empirically test the predictability of MA indicator,RSI indicator and OBV indicator to two index returns from both in-sample and out-sample,the Bootstrap method was used to check the robustness of the result.The results show that technical indicators have the ability to predict market returns.According to the technology analysis trading rules can get better gain of investment,and the prediction ability of technical indicators is stronger than that of macroeconomic indicators.Second,we divide the market ranging from2005 to 2016 into four states which is high sentiment and high attention,high sentiment and low attention,low sentiment and high attention,low sentiment and low attention,then we empirically examine the forecasting ability of technical indicators on market index returns in different states.The result show that when investor sentiment is high,the higher the attention is,the more predictable the technical analysis is,when the investor's sentiment is low,the lower the attention,the more predictable the technical analysis is.Third,using A-share listed company's shareholding quarterly data between 2005 and 2016 by institutional investors,we study the trading behavior of the institutional investor to the company's stock in the case of a rise or fall in the stock price and the impact of this behavior on the company's stock price trend.The research finds that institutional investors are momentum traders,the smaller shareholding on the company the institutionalinvestors have,the stronger momentum trading behavior of institutional investors is and the more obvious the company's stock price trend is,using moving average timing strategy can gain more return.Fourth,by using A-share listed company's quarterly profit index data between 2005 and 2016,we have studied the influence of the company's earnings change on the stock price trend,the results show that when a company's profit index has sustained highly positive(negative)growth,its stock price will be easy to form trend,more excessive return can be obtained by using moving average timing strategy.The cause of this phenomenon may be that investors exist linear extrapolation of the company's profitability based on the "law of small numbers",this linear extrapolation will reduce investors' uncertainty about the stock price and cause overconfidence.Investors' overconfidence can make stock price deviate from the basic value in a short time.In the following time,with the openness of information,stock price will return to its basic value,which will generate a trend.Fifth,using A-share listed company's s daily frequency close price data between 2005 and 2016,we construct a stock trend factor on the use of normalized moving average indicators,the results show that the trend factor has the ability to predict the cross section returns of the stocks.After risk adjustment,there is still a significant alpha gain.By the way of constructing value factors in the Fama-French three factor model,we construct a trend factor return and bring it to asset pricing model.It is found that the model is more suitable for China's stock market.