Document Type

Journal Article

Department/Unit

Department of Geography

Title

Numerical analysis of the statistical properties of uniform design in stated choice modelling

Language

English

Abstract

Stated choice methods have been widely used in transportation studies since 1980s. In recent years, much research attention has been paid to develop optimal or efficient designs for choice experiments, such as the so-called D-optimal design, which does not seek for orthogonality as the traditional approach does but aims at minimizing the determinant of the variance-covariance matrix of the parameter estimators. This paper examines the statistical properties of an alternative design method-uniform design, which also does not look for orthogonality but aims at maximizing uniformity-a measure that is closely related to model efficiency. We compare the estimation efficiency and prediction efficiency of uniform design with that of the traditional fractional factorial orthogonal design in stated choice modelling. Monte Carlo experiments are used to generate models, whose parameters vary in scale. The results show that though uniform design uses a lot fewer profiles than orthogonal designs do, its prediction and estimation efficiencies in stated choice modelling are comparable to that of orthogonal design. © 2009 Taylor & Francis.

Publication Date

2009

Source Publication Title

Transport Reviews

Volume

29

Issue

5

Start Page

619

End Page

634

Publisher

Taylor & Francis

DOI

10.1080/01441640902829454

Link to Publisher's Edition

http://dx.doi.org/10.1080/01441640902829454

ISSN (print)

01441647

ISSN (electronic)

14645327

This document is currently not available here.

Share

COinS