A Hierarchical Bayesian multidimensional scaling methodology for accommodating both structural and preference heterogeneity

Joonwook Park, Wayne Desarbo, John C. Liechty

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Multidimensional scaling (MDS) models for the analysis of dominance data have been developed in the psychometric and classification literature to simultaneously capture subjects' preference heterogeneity and the underlying dimensional structure for a set of designated stimuli in a parsimonious manner. There are two major types of latent utility models for such MDS models that have been traditionally used to represent subjects' underlying utility functions: the scalar product or vector model and the ideal point or unfolding model. Although both models have been widely applied in various social science applications, implicit in the assumption of such MDS methods is that all subjects are homogeneous with respect to their underlying utility function; i.e., they all follow a vector model or an ideal point model. We extend these traditional approaches by presenting a Bayesian MDS model that combines both the vector model and the ideal point model in a generalized framework for modeling metric dominance data. This new Bayesian MDS methodology explicitly allows for mixtures of the vector and the ideal point models thereby accounting for both preference heterogeneity and structural heterogeneity. We use a marketing application regarding physicians' prescription behavior of antidepressant drugs to estimate and compare a variety of spatial models.

Original languageEnglish (US)
Pages (from-to)451-472
Number of pages22
JournalPsychometrika
Volume73
Issue number3
DOIs
StatePublished - Sep 1 2008

Fingerprint

Social Sciences
Marketing
Psychometrics
Antidepressive Agents
Prescriptions
Scaling
Physicians
Methodology
Model
Utility Function
Spatial Model
Social sciences
Unfolding
Scalar, inner or dot product
Drugs

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Applied Mathematics

Cite this

@article{a73d2d1dffa64751a53cf6372f2dd2df,
title = "A Hierarchical Bayesian multidimensional scaling methodology for accommodating both structural and preference heterogeneity",
abstract = "Multidimensional scaling (MDS) models for the analysis of dominance data have been developed in the psychometric and classification literature to simultaneously capture subjects' preference heterogeneity and the underlying dimensional structure for a set of designated stimuli in a parsimonious manner. There are two major types of latent utility models for such MDS models that have been traditionally used to represent subjects' underlying utility functions: the scalar product or vector model and the ideal point or unfolding model. Although both models have been widely applied in various social science applications, implicit in the assumption of such MDS methods is that all subjects are homogeneous with respect to their underlying utility function; i.e., they all follow a vector model or an ideal point model. We extend these traditional approaches by presenting a Bayesian MDS model that combines both the vector model and the ideal point model in a generalized framework for modeling metric dominance data. This new Bayesian MDS methodology explicitly allows for mixtures of the vector and the ideal point models thereby accounting for both preference heterogeneity and structural heterogeneity. We use a marketing application regarding physicians' prescription behavior of antidepressant drugs to estimate and compare a variety of spatial models.",
author = "Joonwook Park and Wayne Desarbo and Liechty, {John C.}",
year = "2008",
month = "9",
day = "1",
doi = "10.1007/s11336-008-9064-1",
language = "English (US)",
volume = "73",
pages = "451--472",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "3",

}

A Hierarchical Bayesian multidimensional scaling methodology for accommodating both structural and preference heterogeneity. / Park, Joonwook; Desarbo, Wayne; Liechty, John C.

In: Psychometrika, Vol. 73, No. 3, 01.09.2008, p. 451-472.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Hierarchical Bayesian multidimensional scaling methodology for accommodating both structural and preference heterogeneity

AU - Park, Joonwook

AU - Desarbo, Wayne

AU - Liechty, John C.

PY - 2008/9/1

Y1 - 2008/9/1

N2 - Multidimensional scaling (MDS) models for the analysis of dominance data have been developed in the psychometric and classification literature to simultaneously capture subjects' preference heterogeneity and the underlying dimensional structure for a set of designated stimuli in a parsimonious manner. There are two major types of latent utility models for such MDS models that have been traditionally used to represent subjects' underlying utility functions: the scalar product or vector model and the ideal point or unfolding model. Although both models have been widely applied in various social science applications, implicit in the assumption of such MDS methods is that all subjects are homogeneous with respect to their underlying utility function; i.e., they all follow a vector model or an ideal point model. We extend these traditional approaches by presenting a Bayesian MDS model that combines both the vector model and the ideal point model in a generalized framework for modeling metric dominance data. This new Bayesian MDS methodology explicitly allows for mixtures of the vector and the ideal point models thereby accounting for both preference heterogeneity and structural heterogeneity. We use a marketing application regarding physicians' prescription behavior of antidepressant drugs to estimate and compare a variety of spatial models.

AB - Multidimensional scaling (MDS) models for the analysis of dominance data have been developed in the psychometric and classification literature to simultaneously capture subjects' preference heterogeneity and the underlying dimensional structure for a set of designated stimuli in a parsimonious manner. There are two major types of latent utility models for such MDS models that have been traditionally used to represent subjects' underlying utility functions: the scalar product or vector model and the ideal point or unfolding model. Although both models have been widely applied in various social science applications, implicit in the assumption of such MDS methods is that all subjects are homogeneous with respect to their underlying utility function; i.e., they all follow a vector model or an ideal point model. We extend these traditional approaches by presenting a Bayesian MDS model that combines both the vector model and the ideal point model in a generalized framework for modeling metric dominance data. This new Bayesian MDS methodology explicitly allows for mixtures of the vector and the ideal point models thereby accounting for both preference heterogeneity and structural heterogeneity. We use a marketing application regarding physicians' prescription behavior of antidepressant drugs to estimate and compare a variety of spatial models.

UR - http://www.scopus.com/inward/record.url?scp=52949133407&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=52949133407&partnerID=8YFLogxK

U2 - 10.1007/s11336-008-9064-1

DO - 10.1007/s11336-008-9064-1

M3 - Article

VL - 73

SP - 451

EP - 472

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 3

ER -