FORECASTING URBAN BUS TRANSIT DAILY REVENUE (

This paper develops three mathematical employment-sensitive revenue models for the purpose of forecasting urban bus transit daily revenues. More specifically, this paper develops and estimates 1) a non-linear model, 2) a Box-Cox model in which both the dependent and independent variables are subjected to different power transformations, and 3) a Box-Cox model in which only independent variables are subjected to power trans- formation. The developed models are estimated using cross-sectional route-level-based data, collected from various public agencies in Amman, Jordan. A number of policy variables representing bus transit services, land use, and area-wide socio-economic characteristics were included in each model. Segmentation of the total district employment was advantageous in determining the contribution of each employment type to specific bus transit revenue.