logo_bandeau

A new shifted log-normal distribution for mitigating 'exploding' implicit prices in mixed multinomial logit models

Auteur: 
Romain CRASTES DIT SOURD
Résumé: 

This paper introduces a new shifted negative log-normal distribution for the price parameter in Mixed Multinomial logit models. The new distribution, labelled as the shifted negative log-normal distribution, has desirable properties for welfare analysis and in particular a point-mass which is further away from zero than the negative log-normal distribution. This contributes to mitigating the 'exploding' implicit prices issue commonly found when the price parameter is speci ed as negative log-normal and the model is in preference space. The new distribution is tested on 10 stated preference datasets. Comparisons are made with standard alternative approaches such as the willingness-to-pay space approach. It is found that the new -shifted distribution yields much lower mean marginal WTP estimates compared to the negative lognormal specification (up to 99% lower) and similar to the values derived from a multinomial logit while at the same time tting the data as well as the negative lognormal speci cation and much better than the willingness-to-pay space approach.
 

Lieu: 

En salle C216

Catégorie: 
Date de début: 
Jeudi 28 octobre 2021 - 12:30
Date de fin: 
Jeudi 28 octobre 2021 - 14:30