To finely grasp the curb parking demand pattern at a micro-spatial scale,quantile regression models for two typical scenarios,namely workdays and weekends were proposed,to explore the nonlinear effects of land utilization,curb parking spaces,traffic factors,and socioeconomic and population on curb parking space occupancy.The results show that quantile regression is superior to linear regression in capturing the complex relationship between variables.And the coefficients and significance levels of explanatory variables change as the quantile changes.At the same time,there is an obvious heterogeneity between workdays and weekends.For example,the number of bus stations around the parking sites is closely related to the low and high occupancies,but this variable has no statistical significance under the medium occupancy.The catering services POIs do not significantly influence curb parking space occupancy on workdays.
关键词
交通运输系统工程/路内停车泊位/使用率/分位数回归模型/非线性影响
Key words
engineering of communication and transportation system/curb parking/occupancy/quantile regression/non-linear impact
维普
万方数据