A stochastic multidimensional scaling vector threshold model for the spatial representation of "pick any/n" data

Wayne Desarbo, Jaewun Cho

Research output: Contribution to journalArticle

32 Citations (Scopus)

Abstract

This paper presents a new stochastic multidimensional scaling vector threshold model designed to analyze "pick any/n" choice data (e.g., consumers rendering buy/no buy decisions concerning a number of actual products). A maximum likelihood procedure is formulated to estimate a joint space of both individuals (represented as vectors) and stimuli (represented as points). The relevant psychometric literature concerning the spatial treatment of such binary choice data is reviewed. The nonlinear probit type model is described, as well as the conjugate gradient procedure used to estimate parameters. Results of Monte Carlo analyses investigating the performance of this methodology with synthetic choice data sets are presented. An application concerning consumer choices for eleven competitive brands of soft drinks is discussed. Finally, directions for future research are presented in terms of further applications and generalizing the model to accommodate three-way choice data.

Original languageEnglish (US)
Pages (from-to)105-129
Number of pages25
JournalPsychometrika
Volume54
Issue number1
DOIs
StatePublished - Mar 1 1989

Fingerprint

Threshold Model
Scaling
Carbonated Beverages
Nonlinear Dynamics
Psychometrics
Binary Choice
Probit
Maximum likelihood
Conjugate Gradient
Estimate
Rendering
Maximum Likelihood
Methodology
Model
Direction compound
Datasets

All Science Journal Classification (ASJC) codes

  • Psychology(all)
  • Applied Mathematics

Cite this

@article{97b16c980ccc4f298259352cede826f7,
title = "A stochastic multidimensional scaling vector threshold model for the spatial representation of {"}pick any/n{"} data",
abstract = "This paper presents a new stochastic multidimensional scaling vector threshold model designed to analyze {"}pick any/n{"} choice data (e.g., consumers rendering buy/no buy decisions concerning a number of actual products). A maximum likelihood procedure is formulated to estimate a joint space of both individuals (represented as vectors) and stimuli (represented as points). The relevant psychometric literature concerning the spatial treatment of such binary choice data is reviewed. The nonlinear probit type model is described, as well as the conjugate gradient procedure used to estimate parameters. Results of Monte Carlo analyses investigating the performance of this methodology with synthetic choice data sets are presented. An application concerning consumer choices for eleven competitive brands of soft drinks is discussed. Finally, directions for future research are presented in terms of further applications and generalizing the model to accommodate three-way choice data.",
author = "Wayne Desarbo and Jaewun Cho",
year = "1989",
month = "3",
day = "1",
doi = "10.1007/BF02294452",
language = "English (US)",
volume = "54",
pages = "105--129",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "1",

}

A stochastic multidimensional scaling vector threshold model for the spatial representation of "pick any/n" data. / Desarbo, Wayne; Cho, Jaewun.

In: Psychometrika, Vol. 54, No. 1, 01.03.1989, p. 105-129.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A stochastic multidimensional scaling vector threshold model for the spatial representation of "pick any/n" data

AU - Desarbo, Wayne

AU - Cho, Jaewun

PY - 1989/3/1

Y1 - 1989/3/1

N2 - This paper presents a new stochastic multidimensional scaling vector threshold model designed to analyze "pick any/n" choice data (e.g., consumers rendering buy/no buy decisions concerning a number of actual products). A maximum likelihood procedure is formulated to estimate a joint space of both individuals (represented as vectors) and stimuli (represented as points). The relevant psychometric literature concerning the spatial treatment of such binary choice data is reviewed. The nonlinear probit type model is described, as well as the conjugate gradient procedure used to estimate parameters. Results of Monte Carlo analyses investigating the performance of this methodology with synthetic choice data sets are presented. An application concerning consumer choices for eleven competitive brands of soft drinks is discussed. Finally, directions for future research are presented in terms of further applications and generalizing the model to accommodate three-way choice data.

AB - This paper presents a new stochastic multidimensional scaling vector threshold model designed to analyze "pick any/n" choice data (e.g., consumers rendering buy/no buy decisions concerning a number of actual products). A maximum likelihood procedure is formulated to estimate a joint space of both individuals (represented as vectors) and stimuli (represented as points). The relevant psychometric literature concerning the spatial treatment of such binary choice data is reviewed. The nonlinear probit type model is described, as well as the conjugate gradient procedure used to estimate parameters. Results of Monte Carlo analyses investigating the performance of this methodology with synthetic choice data sets are presented. An application concerning consumer choices for eleven competitive brands of soft drinks is discussed. Finally, directions for future research are presented in terms of further applications and generalizing the model to accommodate three-way choice data.

UR - http://www.scopus.com/inward/record.url?scp=0039263131&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0039263131&partnerID=8YFLogxK

U2 - 10.1007/BF02294452

DO - 10.1007/BF02294452

M3 - Article

VL - 54

SP - 105

EP - 129

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 1

ER -