摘要
在重庆市政府回购、租赁高速公路的背景下,研究了如何合理、准确地预测高速公路交通量,从而为政府部门及高速公路投资者的投资决策提供依据。根据高速公路年交通量样本小、预测期长、受经济因素影响等特点,选用了支持向量机回归来进行多因素单目标的预测。在预测过程中,为了提高精度,首先将所搜集的经济因素进行主成分分析,对指标进行了约减;然后用PSO方法对支持向量机参数进行了优化。
In the background of the government repurchasing the toll highway, this paper studies how to forecast the traffic volume precisely. Due to the character of the samples, SVM (support vector machine) is chosen to deal with this problem. And in order to forecast the traffic volume accurately, the author firstly reduces the economic factors by principal components analysis (PCA). Secondly, particle swarm optimization (PSO) is used to search for the best parameters for the support vector machine (SVM).
出处
《管理评论》
CSSCI
北大核心
2011年第12期32-37,67,共7页
Management Review
基金
教育部人文社会科学研究计划项目(10XJA630010)