摘要
灰色神经网络在人工智能预测领域已经得到广泛的应用,但由于其自身存在局部最小化和收敛速度慢等问题,使其预测精度受到一定的限制。针对其不足,本文提出一种利用粒子群算法优化BP神经网络的学习算法,在此基础上,利用灰色预测方法对股指期货历史数据进行初步预测,并且把初步预测的结果作为优化BP神经网络的输入进行训练和预测,构建了基于粒子群优化的灰色神经网络组合预测模型(PSO-GMNN)。仿真实验结果表明,新预测模型的预测精度高于BP神经网络、灰色神经网络和灰色预测模型,同时也表明了该方法的有效性和可行性。
Gray neural network in the field of artificial intelligence prediction has been applied widely,but it has such problems as the slow speed of convergence,and local minimum,so its forecast precision is limited partly.This paper,in view of its defects,proposes the learning algorithm of the BP neural network optimized by PSO（Particle swarm algorithm）.On the basis of this algorithm,grey prediction is used to make a preliminary forecast for the stock index futures＇ historical data,and the results of initial forecasts are used as the input of the optimized BP neural network to be forecast and trained.A PSO-based Combined forecasting Grey Neural Network model（PSO-GMNN） is built.Finally,the simulation experiment result indicates that the prediction accuracy of the new prediction model is higher than that of the BP neural network,the gray neural network and the gray prediction model.It also shows the effectiveness and feasibility of the method.
出处
《计算机工程与科学》
CSCD
北大核心
2012年第2期146-149,共4页
Computer Engineering & Science
基金
河南省科技攻关项目(092102210108)
