DEVELOPMENT OF STRUCTURAL EQUATIONS MODELS OF THE DYNAMICS OF PASSENGER TRAVEL DEMAND

Simultaneous equation models are estimated using week-long travel diary data from individuals surveyed six months apart. The models are structured to capture three types of relationships among demand levels for five modes of transportation: (1) autocorrelative relationships, capturing temporal stability or inertia in demand for the same mode at different points in time, (2) synchronous relationships capturing complementarity and competition among different modes at the same point in time, and (3) cross-lagged effects, capturing substitution of one mode by another at a later point in time. Both trip rates and travel times are modeled successfully according to several goodness-of-fit indices.