A multiscale spatial analysis of taxi ridership

Taxi plays a supplement role in sustainable development of urban public transport systems. However, the extent to which the built environment affects taxi ridership at various spatial scales deserves further exploration because understanding the true spatial heterogeneity across a varying scale could be valuable for both global and localized policy decision-makings. In this study, we attempt to analyze and discuss the spatial predictors of taxi ridership by utilizing an multiscale geographically weighted regression (MGWR) model and comparing the model's performance to that of ordinary least square (OLS) and geographically weighted regression (GWR) models. Using the taxi data of Xi'an city, we found that the MGWR model could explain 81.8% of the total taxi ridership fluctuations and allows localized and targeted policy makings to help taxi drivers search for passengers and to improve passengers' taxi-hailing experiences in specific districts.

Automated summary

This study analyzes the spatial predictors of taxi ridership using a multiscale geographically weighted regression (MGWR) model and finds that the model performs well in explaining fluctuations in ridership and aiding in localized policy making.