The Stock Market Prediction via Deep Learning

• 1、State Key Laboratory of Software Engineering, Wuhan University, Wuhan 430072

Abstract：Stock market plays an important role in nowadays financial markets. And stock price volatility prediction is regarded as one of the most attractive and meaningful research issues. The key problem of the stock market prediction is how to design a method to improve the prediction accuracy. Some existing researches on stock market price volatility prediction have pointed out that architectures are applied in many existing machine learning algorithms including neural networks with only one hidden layer and support vector machine are using shallow architectures. Psychology results shows with limited samples and finite computing units, those shallow architectures are incapable of representing the complex function, and place restriction on the generalization capability of classifying complicated issues, especially for the rich sensory input. Deep learning achieves the approximation of complex function，characterization of the input data by learning a deep nonlinear network, and shows the great power in extracting the intrinsic feature of the training data. To validate the performance of Deep Learning, we take experiments on HKEx 2001 stock market datasets. The results show that deep learning has a the great power in extracting the intrinsic feature of the training data and it can refect the intrinsic feature of the training data.

Keywords： Deep Learning Stock Market Prediction Machine Learning Multi-Sourcedata Processing