摘要
研究基金净值预测问题,基金净值的变化受到政治、经济等多种因素的影响,变化规律相当复杂,各种影响因素间呈复杂的非线性关系,针对传统的预测方法不能很好反映这种非线性规律,导致基金净值预测精度不高,为了提高基金净值预测精度,提出了一种采用粒子群优化与BP神经网络相结合的组合模型,用于基金净值预测研究。利用粒子群具有较快的收敛速度和优异的全局寻优能力,对BP神经网络的权值和阈值进行仿真,证明提高了基金净值的预测精度。通过实证研究,表明了组合模型比其它模型具有更高的精确度和更快的收敛速度,能准确地捕捉基金净值的变化状况,为基金净值预测提供了一个实用的方法。
Research on the fund net value prediction.Many factors can change the fund net value,such as political,economic and other factors,The fund net value change rule is very complex,there are complicated nonlinear relations among the various factors,and the traditional forecasting method cannot reflect the nonlinear rule,which leads to low prediction accuracy.In order to improve the prediction accuracy of fund net value,we put forward a hybrid model based on particle swarm optimization and the BP neural network,using particle swarm with higher convergence speed and outstanding global optimization ability to improve the prediction precision of fund net value by optimized the weights and threshold of BP neural network.The empirical study shows that the combined model has higher accuracy and faster convergence than other models,can accurately capture the change trend of the fund net value,and is a practical prediction method in fund net value.
出处
《计算机仿真》
CSCD
北大核心
2011年第5期354-357,共4页
Computer Simulation
关键词
组合模型
粒子群
神经网络
基金净值
Hybrid model
Particle swarm
Neural network
Fund net value
