ESTIMATING ELASTICITIES FOR PUBLIC TRANSPORT DEMAND IN ATHENS, GREECE: A TIME SERIES ANALYSIS

This paper describes how research on the demand side of public transportation systems that uses time series data frequently shows conflicting results with respect to fare elasticities and to the factors affecting the demand for transit. In this analysis we complement prior research by developing parsimonious Seemingly Unrelated Regression Equation models with monthly data (from 01/1995 to 04/2001) for a city served by four different modes of public transportation. The results indicate that urban public transport demand in Athens, Greece, is inelastic with respect to fares and fuel prices. Further, different modes of public transport have different fare elasticities, a finding with very important practical implications.