Estimation of Swiss railway demand with computation of elasticities

We adapt the ‘Almost Ideal Demand System’ (AIDS) from consumer expenditure theory to estimate an equation explaining demand for rail passsenger travel. The motivation for using an AIDS is essentially empirical: (1) The model is flexible enough to easily enable the insertion of specific variables to take account of variation in the characteristics of goods (examples cited include ‘speed of transportation’ and ‘density of the railway network’); (2) the obtained equation ensures a good fit to the data and compares favorably with other specifications. We derive a demand function and formulas for the elasticities and other parameters. The model is applied to Swiss data, using a three-way error components model on cross-section and time series data. We apply a linear, simplified, version of the AIDS and we use an iterative approximation of the true AIDS price index. Thanks to bootstrap replications, we simulate the distribution of the most relevant estimates of the elasticities.