Order-invariant prior specification in Bayesian factor analysis

Dennis Leung; Mathias Drton

doi:10.1016/j.spl.2016.01.006

Order-invariant prior specification in Bayesian factor analysis

Dennis Leung, Mathias Drton

University of Washington

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

Using lower triangular loading matrices in Bayesian factor analysis ensures identifiability but may lead to inferences that depend on how the considered variables are ordered. We show how a standard approach to prior specification can be modified to avoid order-dependence.

Original language	English
Pages (from-to)	60-66
Number of pages	7
Journal	Statistics and Probability Letters
Volume	111
DOIs	https://doi.org/10.1016/j.spl.2016.01.006
State	Published - 1 Apr 2016
Externally published	Yes

Keywords

Exploratory factor analysis
Latent factor model
Permutation invariance

Access to Document

10.1016/j.spl.2016.01.006

Cite this

@article{508db5eb0e644b47a1d54b5502e6f579,

title = "Order-invariant prior specification in Bayesian factor analysis",

abstract = "Using lower triangular loading matrices in Bayesian factor analysis ensures identifiability but may lead to inferences that depend on how the considered variables are ordered. We show how a standard approach to prior specification can be modified to avoid order-dependence.",

keywords = "Exploratory factor analysis, Latent factor model, Permutation invariance",

author = "Dennis Leung and Mathias Drton",

note = "Publisher Copyright: {\textcopyright} 2016 Elsevier B.V.",

year = "2016",

month = apr,

day = "1",

doi = "10.1016/j.spl.2016.01.006",

language = "English",

volume = "111",

pages = "60--66",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Order-invariant prior specification in Bayesian factor analysis

AU - Leung, Dennis

AU - Drton, Mathias

PY - 2016/4/1

Y1 - 2016/4/1

N2 - Using lower triangular loading matrices in Bayesian factor analysis ensures identifiability but may lead to inferences that depend on how the considered variables are ordered. We show how a standard approach to prior specification can be modified to avoid order-dependence.

AB - Using lower triangular loading matrices in Bayesian factor analysis ensures identifiability but may lead to inferences that depend on how the considered variables are ordered. We show how a standard approach to prior specification can be modified to avoid order-dependence.

KW - Exploratory factor analysis

KW - Latent factor model

KW - Permutation invariance

UR - http://www.scopus.com/inward/record.url?scp=84956668655&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2016.01.006

DO - 10.1016/j.spl.2016.01.006

M3 - Article

AN - SCOPUS:84956668655

SN - 0167-7152

VL - 111

SP - 60

EP - 66

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

ER -

Order-invariant prior specification in Bayesian factor analysis

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this