01595640

Component

A simulation evaluation is presented to compare alternative estimation techniques for a five-alternative MNP with random parameters, including cross-sectional and panel datasets and for scenarios with and without correlation among random parameters. Results suggest that the MACML approach provided the best performance in terms of the accuracy and precision of parameter recovery and estimation time. The GHK approach with Halton sequences, when combined with the CML approach for panel datasets, appears to be a good contender for MNP estimation, albeit with longer estimation time than the MACML approach. The sparse grid approach did not perform well in recovering the parameters as the dimension of integration increased, particularly so with the panel datasets. The MCMC approach performed well in datasets without correlations among random parameters, but exhibited limitations in datasets with correlated parameters.

This paper was sponsored by TRB committee ADB40 Standing Committee on Transportation Demand Forecasting. Alternate title: Simulation Evaluation of Emerging Estimation Techniques for Mixed Multinomial Probit Models.

01584066

16-5139

English

0000-0002-0715-8121

25p

2016

Transportation Research Board 95th Annual Meeting

Digital/other

References; Tables

Estimating; Markov chains; Multinomial probits; Simulation

Planning and Forecasting; Transportation (General)

Transportation Research Board Annual Meeting 2016 Paper #16-5139

TRIS, TRB, ATRI

Jan 12 2016 6:14PM