基于人工神经网络的股市预测模型

建立了构成基于人工神经网络的3种股市预测模型（基本数据模型、技术指标模型和宏观分析模型）,分析了神经网络应用于股市预测的实效性。实证分析表明,3种模型对上证综合指数的拟合效果均较好。在“基本数据模型“中,建立带有附加动量项和自适应学习速率的BP网络,具有较快的运算速度和逼近性能。在“技术指标模型”中,通过一些股市重要技术指标的引入,使其增加了反映市场各方面深层内涵的信息,而且网络的泛化能力有所提高。在“宏观分析模型”中,引入了影响股市的5项主要宏观经济指标,使模型包含了宏观经济基本面的更多信息,强化了股市神经网络模型的应用价值。

Forecast model for stock market based on artificial neural networks

A series of stock market forecast models were provided, based on the techniques of Artificial Neural Networks(ANN). They are "Basic Data Model","Technological Index Model" and "Macro-analysis Model". For the nature of ANN, such as self-organization, adaptability , self-learning and robustness, these models prove to be effective and applicable for the forecast of stock market.