Financial time series has high randomicity and nonlinearity. Neural network is quite suitable in the process of financial time series data for its good ability of nonlinear mapping and generalization.
金融时间序列具有很强的随机性和非线性性，而神经网络具有良好的非线性映射能力及自适应、自学习和良好的泛化能力，因此非常适合处理金融时间序列这样的数据。
The BP neural network is adopted to realize the reinforcement learning to strengthen the generalization ability.
应用BP神经网络实现强化学习，以增强系统的泛化能力；
Meanwhile some research achievements had been summarized about approaching and generalization ability of the multilayer forward neural network, which was rationale for the next modeling.
同时总结了有关多层前向网络逼近能力和泛化能力的一些研究成果，作为后续建模工作的理论依据。
应用推荐