| 基于多元回归分析的事件持续时间预测 |
| |
| 引用本文： | 刘伟铭,管丽萍,尹湘源.基于多元回归分析的事件持续时间预测[J].公路交通科技,2005,22(11):126-129. |
| |
| 作者姓名： | 刘伟铭 管丽萍 尹湘源 |
| |
| 作者单位： | 1. 华南理工大学,广东,广州,510641
2. 浙江万里学院,浙江,宁波,315100 |
| |
| 摘 要： | 事件持续时间预测是实现先进的事件管理系统的一个重要基础。本文通过多元回归分析法对高速公路事件持续时间预测问题进行研究。首先通过方差分析(ANOVA)确定对事件持续时间具有显著影响的因素。在此基础上,通过逐步回归分析确定用于高速公路事件持续时间预测的最佳变量组合并建立多元线性回归模型。经检验,预测值与实际值的相关系数为0.8573,预测结果基本能够反映真实的事件持续时间情况。
|
| 关 键 词： | ITS 事件管理 事件持续时间 逐步回归分析 方差分析 |
| 文章编号： | 1002-0268(2005)11-0126-04 |
| 收稿时间： | 2004-07-28 |
| 修稿时间： | 2004年7月28日 |
Prediction of Incident Duration Based on Multiple Regression Analysis |
| |
| LIU Wei-ming,GUAN Li-ping,YIN Xiang-yuan.Prediction of Incident Duration Based on Multiple Regression Analysis[J].Journal of Highway and Transportation Research and Development,2005,22(11):126-129. |
| |
| Authors: | LIU Wei-ming GUAN Li-ping YIN Xiang-yuan |
| |
| Institution: | 1. College of Traffic and Communications, South Chain University of Technology, Guangdong Guangzhou 510641, China; 2.Zhejiang Wanli University, Zhejiang Ningbo 315100, China |
| |
| Abstract: | The prediction of incident duration is an important foundation of expressway incident management system.In this paper,the problem of expressway incident duration prediction is studied.First,the marked influence factors of incident duration are determined by analysis of variance(ANOVA).Then,stepwise regression analysis is used to select a best group of factors for the prediction of expressway incident duration,and the multiple linear regression model is established.Finally,the prediction effect of the regression model is tested.The test result indicates that the correlation of the prediction values and factual values is 0.8573,so the test result can generally represent actual incident duration. |
| |
| Keywords: | Intelligent Transportation System Incident management Incident duration Analysis of variance Stepwise regression analysis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
