Short term prediction of stock market prices with joint empirical modal decomposition and recurrent neural networks

An EMD neural network model for modeling and forecasting the Chinese stock market is proposed by applying EMD algorithm, chaos analysis and neural network theory. Firstly, the EMD decomposition algorithm is applied to decompose the original stock market time series into basic modal components of different scales, and further analyzed on the basis of this, which shows that there are chaotic characteristics in the Chinese stock market; and then the combination of chaos analysis and neural networks is used for prediction, which improves the model's ability to learn from a variety of objective functions, and effectively improves the prediction accuracy. Experiments show that the method has high accuracy compared with existing methods.