Data Analysis of Roadway Attributes through Partial Least Squares Regression

摘要：

Based on the actual data collected from black spots on mountain highways,this paper investigated the influences of roadways attributes upon vehicle speed.The significance of different factors contribute to vehicle speed were represented by the modified coefficients of roadway attributes,and a model was established based on the modified coefficients.Moreover,we took advantage of Logarithmic regression and then applied Partial Least Squares regression for computing the modified coefficients of roadway attributes towards vehicle speed.Through analysis up model,some favorable results obtained.To fix the invalidity of classic least squares regression on this issue,we present the feasible approach to establish regression model.

关键词： coefficient Partial Least Squares Regression

作者: Weiguo Li Hanjie Zhang Xiaoping Du Kun Qian Cuiying Li

作者单位: Key Laboratory of Road Safety Technologies,Ministry of Communications,P.R.China Beihang University,Beijing,P.R.China

会议类型: 国际会议

会议名称: 2010 2nd IEEE International Conference on Information and Financial Engineering(2010年第二届IEEE信息与金融工程国际会议 ICIFE 2010)

会议地点: 重庆

会议语种:英文

页码: 466-468

在线出版日期: 2010-09-17（万方平台首次上网日期，不代表论文的发表时间）