Factors affecting public transportation usage rate: Geographically weighted regression

摘要: As the number of private vehicles grows worldwide, so does air pollution and traffic congestion, which typically constrain economic development. To achieve transportation sustainability continued development, dependency on must be decreased by increasing public usage. However, without knowing key factors that affect usage, developing strategies effectively improve usage is impossible. Therefore, this study respectively applies global local regression models to identify rates for 348 regions (township or districts) in Taiwan. The model, Tobit model (TRM), used estimate one set parameters are associated with explanatory variables explain regional differences rates, while geographically weighted (GWR), estimates differently depending spatial correlations among neighbouring regions. By referencing related studies, 32 potential four categories, social-economic, land use, transportation, chosen. Model performance compared terms mean absolute percentage error (MAPE) autocorrelation coefficient (Moran’ I). Estimation results show GWR has better prediction accuracy accommodation autocorrelation. Seven significantly tested, most have differ across Based these findings, proposed