Using nonparametric methods, the paper examines the specification of a model to evaluate the distribution of willingness-to-pay (WTP) for travel time savings from binomial choice data from a simple time-cost trading experiment involving four public transport modes. A formulation in preference space in terms of constant marginal utilities is rejected in favor of a formulation in log(WTP) space for each mode. Results further indicate that the log(WTP) distribution can be decomposed into an independent random variable and a linear index shifting the location of the log(WTP) distribution, which is useful for parametric modeling. The index strongly indicates that small time savings are valued less than large time savings for all four modes in this experimental choice situation. The sign for journey duration varies significantly between modes, which could be due to self-selection into modes. For the covering abstract please see ITRD E135207.
Samenvatting