Based On The Main Capital Flow Direction Of The SSE Index Forecasting Scheme Design Research

The stock market can be defined as such a market.On the one hand,it provides facilities for companies that need financing;on the other hand,it provides opportunities for investors who need to invest.By forecasting the rise and fall of the stock index,it can provide guidance for individuals and enterprises when they enter the financial market,and it can also provide theoretical significance for the formulation of the government’s economic policy.However,the stock market is a complex system filled with various kinds of information.It is not only affected by the information in the past,but also affected by the current political,economic,and psychological factors.Therefore,it is difficult for the stock index to change accurately.At present,the method for forecasting the stock index ups and downs is mainly applied technology analysis method and measurement time series analysis method.Among them,the application technology method uses more groups because it almost does not need too much analysis but is based on personal investment habits and experience.Subjective colors are strong.The time series method of measurement is an effective method that is used in an ideal situation.It requires that the input independent variable and the target variable have a linear relationship.If it is a nonlinear case,the result has no reference significance.Under such development conditions,popular machine learning algorithms have gained widespread attention and applications from investors.One of the commonly used artificial intelligence method is support vector machine(SVM).Support vector machine technology in its classification,regression and other aspects of the excellent features successfully used in machine learning,pattern recognition and other fields.It is because of these characteristics of SVM that it provides a new way of thinking for the prediction of the change of stock index.This article combines the main cash flow model,support vector machine as a tool to build a stock index change forecasting program.And this article takes the Shanghai Composite Index as the research object,through the comparative method,selects 41 technical indicators(excluding the main capital flow indicators)or 42 technical indicators(including the main capital flow indicators)as input variables,the stock index up and down classification As a target classification variable,PCA-SVM was modeled by using R language and a reasonable empirical result was obtained.The conclusion shows that the forecasting accuracy rate of the main cash flow index is higher than the forecasting accuracy rate of the main cash flow index.Finally,for the sake of the completeness of the result,this paper also divides the forecast period into rising markets,falling markets and shock cities.Among them,the forecast of rising cities is the best,and the forecast of falling markets is the worst.Taken together,the combination of the main cash flow model and the machine learning method has certain guiding significance for the stock market forecast and investment.This article also provides a convenient and practical solution for investors and government regulation.