As one of the most important financial instruments,stock plays an irreplaceable role in optimizing resource allocation,long-term capital financing and so on.The stock price has the characteristics of high noise,nonlinear,easy to be affected by the policy,so it is difficult for investors to avoid the risk.The strong nonlinear processing ability of artificial neural network is suitable to solve the problem of stock forecasting,this paper uses the neural network to study the stock price forecasting model.The main contents are summarized as follows:The stock index forecasting model has high noise and nonlinearity,which is easy to be affected by the policy and so on.Therefore,the traditional statistical analysis method can not meet the accuracy of the prediction error.Artificial neural network(ANN)is widely used in the field of prediction because of its strong nonlinear modeling ability,self-learning ability and fault tolerance ability.The intelligent optimization algorithm is introduced to improve the prediction accuracy of the algorithm.Therefore,this paper uses the improved artificial neural network to predict the Shanghai Composite Index closing price.The main contents and innovations of this paper are summarized as follows:1.A GA-BP neural network prediction model based on time series is established.Firstly,the input and output indexes in the neural network are determined by time series,and then the genetic algorithm is used to optimize the connection weight and threshold of BP neural network,Through the analysis of the forecast results of Yunnanbaiyao's stock price,the model has better prediction performance.2.A new stock price forecasting model which combine improved particle swarm optimization algorithm(IPSO)with BP neural network is established.Firstly,the non-exponential decreasing inertia weight,asynchronous learning factor and adaptive mutation operator are used to improve the standard particle swarm.Then,the improved particle swarm optimization algorithm is used to optimize the weight and threshold value of BP neural network connection.Through the simulation experiment of Shanghai Stock Index closing price show that the convergence speed of the model is high and the error value of the prediction is low,which indicates that the improved effect of the algorithm is effective.3.The Elman dynamic neural network stock price forecasting model based on principal component analysis is established.First of all,the principal component analysis method is used to preprocess the stock price index,which simplifies the calculation process.Then,the Elman dynamic neural network is used to simulate the Shanghai Composite Index,and the experimental results show the superiority of the model. |