David Palma - Extending the Multiple Discrete Continuous (MDC) modelling framework to consider complementarity, substitution, and an unobserved budget

There are many tools available for demand forecasting at the individual level: discrete choice models to forecast what is bought, and linear regression to forecast how much of it is bought. Multiple-Discrete-Continuous (MDC) models can forecast both what and how much is purchased simultaneously, but they have historically being limited by a lack of complementarity and substitution effects, as well as the need to define a budget. The model presented in this paper extends the traditional MDC framework to allow for explicit and robust complementarity and substitution effects, while also allowing for unobserved budgets. With a freely available computational implementation for the R programming language, the eMDC model becomes an attractive and accessible new tool for demand estimation.

Palma, D. and Hess, S. (2002) Extending the Multiple Discrete Continuous (MDC) modelling framework to consider complementarity, substitution, and an unobserved budget. Transportation Research 161B, 13 – 35.
https://doi.org/10.1016/j.trb.2022.04.005