Font Size: a A A

Based On The Average Forecast Of Stock Market Investment Strategy To Build And Empirical

Posted on:2013-08-01Degree:MasterType:Thesis
Country:ChinaCandidate:W TianFull Text:PDF
GTID:2249330374985662Subject:Finance
Abstract/Summary:PDF Full Text Request
For a long time, the focus of the stock investment strategy building has been ontwo aspects: one is just transactions with the moving average rules. The ultimate aim isoften not for investment, but in order to test the market validity. The other is to build aportfolio for practical using. And recent studies on the investment strategy of the stockis turned to the strategy based on the stock market prediction.This paper builds a strongly operational investment strategy based on movingaverage trading rules. The key point of the strategy is to get the predictive value of thestock price by the predict model. This is because that moving average has a lagcomparing to stock price. SO it is very important for investors to judge the price trendand inflection by the moving average prediction.First we use two-level-AR model to predict. We use the moving average data toestablish the up level model and then use the moving average and shares data toestablish the lower level model. the MA predictive value of up level model canregulate predictive value of the lower level model, extend its prediction range, whileminimizing the prediction bias.Second we use AC model to predict. AC algorithm is a non-parametric forecastmethod of graphical fitting, which is useful for finding the inflection point of movingaverage system. It gets the predictive valve by the transform and combination of thehistorical data ‘s continuation which are similar to the current period state.The result of simply using the AC algorithm is not satisfactory because that thestock graphics is lack of regularity. So we introduction the EMD to greatly improve theprediction precision and accuracy of the AC algorithm by dividing the curve of theoriginal data into a number of certain regularity curves.On the base of the predictive valve of the three models,we do an empirically studyarbitrarily selecting30stocks. The results show that the large-cap stocks’ has a smallerreturn and volatility than the small-cap stocks’,and the AC model based on the EMDhas the highest prediction accuracy and the best result of a transaction. So theinvestment strategy built on the AC model which based on the EMD has a good reality operability and profitability, and can be used as an important reference for the majorityof stock investors.
Keywords/Search Tags:two-level-AR, analog complexing, empirical mode decomposition, modelimprovements, investment strategy
PDF Full Text Request
Related items