Exploring Spatial Variation of Urban Taxi Ridership Using Geographically Weighted Regression
Exploring Spatial Variation of Urban Taxi Ridership Using Geographically Weighted Regression
复制标题
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Xinwu Qian;S. Ukkusuri
中科院分区：
文献类型：
--
作者：
Xinwu Qian;S. Ukkusuri
出租车是城市交通系统的重要组成部分，其功能和每日服务的乘客量。虽然已进行了广泛的研究，以分析公共汽车和轻轨乘客的交通需求，很少有人努力探索的出租车乘客。在本文中，作者提出了第一个研究，以分析城市出租车乘客的空间格局。利用纽约市大规模出租车出行数据和地理数据库，研究了出租车载客量与人口统计、城市土地利用和其他公共交通方式可达性相关变量之间的关系。地理加权回归（GWR）的城市出租车乘客量建模和可视化的空间分布参数估计。结果表明，GWR完全优于普通最小二乘法（OLS）的拟合优度和解释力。此外，城市形态被发现有显着影响的独立变量的校准和不考虑空间非平稳效应可能会导致有偏的估计。总的来说，这项研究深入了解了出租车乘客量在空间上的变化，其结果可作为预测出租车需求、制定适当的出租车行业法规和制定城市规划的指导。
Taxicab is an important component of the urban transit system in terms of its functionality and the amount of passengers served daily. While extensive studies have been conducted to analyze the transit demand such as bus and light-rail ridership, few efforts are made to explore the taxi ridership. In this paper, the authors present the first study to analyze the spatial pattern of the urban taxi ridership. The relationship between the taxi ridership and variables related to demographics, urban land use and the accessibility to other public transport modes are investigated using large scale New York City (NYC) taxi trip data and geographical database. The geographically weighted regression (GWR) is implemented to model the urban taxi ridership and visualize spatial distributions of parameter estimations. The results suggest that the GWR completely outperforms the ordinary least square (OLS) method in terms of the goodness of fit and explanatory power. Additionally, the urban form is found to have significant impact on the calibration of independent variables and failing to account for the spatial non-stationary effect may lead to biased estimations. In general, the study provides an in-depth understanding on how taxi ridership varies over the space and the results may serve as a guidance for predicting taxi demand, developing proper regulations for the taxi industry and drawing up urban plans.