Accounting for Stochastic Variables in Discrete Choice Models

The estimation of discrete choice models requires collecting data about the socioeconomic characteristics of individuals and measuring the attributes describing the alternatives within each individual’s choice set. Even though some attributes are intrinsically stochastic (e.g. travel times) or are subject to non-negligible measurement errors (e.g. waiting times), they are usually assumed fixed and deterministic. Indeed, even an accurate measurement can be biased as it might differ from the original (experienced) value perceived by the individual. Experimental evidence suggests that discrepancies between the values measured by the modeller and experienced by the individuals can lead to incorrect parameter estimates. On the other hand, there is, as usual, an important trade-off between data quality and collection costs. This paper explores the inclusion of stochastic variables in discrete choice models through an econometric analysis that allows identifying the most suitable specifications. Various model specifications were experimentally tested using Monte Carlo simulation. Comparisons included tests for unbiased parameter estimation, computation of marginal rates of substitution and demand forecasts (response analysis) under the implementation of different transport policies. Results show that error components models can effectively deal with stochastic variables. Also, as in previous misspecification tests reported in the literature, the Multinomial Logit model proved to be quite robust for estimating marginal rates of substitution and forecasting demand for realistic policies, especially when it was estimated with a large number of observations.