APPLICATION OF MODELS BASED ON STATED AND REVEALED PREFERENCE DATA FOR FORECASTING PASSENGER TRAFFIC BETWEEN EAST AND WEST DENMARK

The entry of new travel alternatives into an existing market poses special problems for demand forecasting, specifically, as there are not observed choice data for these alternatives. In the 1990s methods were developed which used both stated preferences (SP) and revealed preference (RP) data simultaneously to estimate the model parameters. Further problems, however, have been identified with the application on these models. This paper describes a three-step estimation procedure that has been developed to take account of these problems and illustrates how the approach has been applied in the development of a forecast model to predict the number of passengers on a new road and rail fixed links, which are being constructed across the Storebaelt strait in Denmark.