Stock Prediction For Trading Advice

Stock market became the most important and absolutely necessary part of securities trade and financial market after the establishment in China. More and more investors pay attention to stock market. So the analysis and forecast of stock market have not only signification of theory but also merit of application. Stock market is a very complex nonlinear dynamic system. Neural network has the capability of approximating any nonlinear system and specialty of self-learning and self-adopting. The experiments show that the method of modeling stock market using neural network has a satisfying result in near-period or middle-period stock prediction.Firstly, the paper states the background, meaning, domestic and international research current situation and basic theories about stock market prediction and analysis various factors that affect the stock market, stock market prediction methods and the problems in predicting. Secondly, the widely used BP neutron network is applied to predict the close price of Microsoft Company stock and the experimental results fit the change tendency of the actual close price.Next, the stock prediction model based on echo state network is proposed on the base of the high-nonlinear specialty of stock market and the shortcoming of basic BP algorithm in the slow convergence speed and local minimum. Similarly, the model is applied to predicting the close price of Microsoft Company stock and the experimental results are compared with that of BP neural networks. According to the MACD strategy, predication values are estimated and valuable trading advice are given.Finally, a stock price prediction simulation system is set up based on the above model. The validity of the model is tested by analyzing six stocks such as Microsoft Company stock and Chinese petroleum chemical stock and so on.The experimental results show that it is applicable and feasible to use neural network to predict stock market. However,the amount of training samples used in our model is relatively small, and the dimension of each sample is low. We do not consider the influence of the government policy and some other factors, which usually play an important role in the fluctuation of the stock market. So there is still a long way to go before applying it to real applications, and need to be improved further.