Research on simple moving average trading system based on SVM

To gain stable and continuous profit return, trading system based on some trade rules are used by more and more investors and institutions while simple moving average system is one of the trading systems. In order to improve the efficiency of the system, this paper set up a Support Vector Machine (SVM) model trained by RAVI and four other kinds of trend technique indicators to reflect the stock price time series into high dimensional characterization space, and filter the signals of 5-60 simple moving average trading system with SVM model. The results of testing on the stock index of Inner Mongolia Yili showed obvious promotion and better evaluation indicators compared to other trading systems and filtering methods. It is feasible approach to improve simple moving average trading system with SVM consequently.