Seminar: “Modelling time allocation with discrete-continuous models”
Speaker: Chiara Calastri (Choice Modelling Centre and Institute for Transport Studies University of Leeds, UK)
The MDCEV modelling framework has established itself as a preferred method for modelling time allocation, with data very often collected through travel or activity diaries. However, standard implementations fail to recognise the fact that many of these datasets contain information on multiple days for the same individual, with possible substitution between days. This paper discusses how the theoretical accommodation of these effects is not straightforward, especially with budget constraints at the day and multi-day level. We instead rely on additive utility functions where we accommodate correlation between
activities at the within-day and between-day level using a mixed MDCEV model, with multi-variate random distributions. We put forward adaptations of the standard Pinjari and Bhat (2010) forecasting approach to allow us to make links across days also in model application. Finally, we illustrate the issue and the methods using two different time use datasets, confirming our theoretical points and highlighting the benefits of allowing for correlation across days in estimation and substitution in forecasting.