A note on identifiability conditions in confirmatory factor analysis

William Leeb

doi:10.1016/j.spl.2021.109190

A note on identifiability conditions in confirmatory factor analysis

William Leeb

Mathematics

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

1 Scopus citations

Abstract

Recently, Chen, Li and Zhang established conditions characterizing asymptotic identifiability of latent factors in confirmatory factor analysis. We give an elementary proof showing that a similar characterization holds non-asymptotically, and prove a related result for identifiability of factor loadings.

Original language	English (US)
Article number	109190
Journal	Statistics and Probability Letters
Volume	178
DOIs	https://doi.org/10.1016/j.spl.2021.109190
State	Published - Nov 2021

Bibliographical note

Funding Information:
I am grateful to Xiaoou Li for discussing her work from ( Chen et al., 2020 ), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of America IIS-1837992 and BSF award 2018230 .

Funding Information:
I am grateful to Xiaoou Li for discussing her work from (Chen et al. 2020), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of AmericaIIS-1837992 and BSF award2018230.

Publisher Copyright:
© 2021 Elsevier B.V.

Keywords

Confirmatory factor analysis
Factor loadings
Identifiability
Latent factors
Matrix factorizations

Access

10.1016/j.spl.2021.109190

OpenUrl availability

Full text

Cite this

@article{65a824bfb6f24f8aa61d683c586056fd,

title = "A note on identifiability conditions in confirmatory factor analysis",

abstract = "Recently, Chen, Li and Zhang established conditions characterizing asymptotic identifiability of latent factors in confirmatory factor analysis. We give an elementary proof showing that a similar characterization holds non-asymptotically, and prove a related result for identifiability of factor loadings.",

keywords = "Confirmatory factor analysis, Factor loadings, Identifiability, Latent factors, Matrix factorizations",

author = "William Leeb",

note = "Funding Information: I am grateful to Xiaoou Li for discussing her work from ( Chen et al., 2020 ), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of America IIS-1837992 and BSF award 2018230 . Funding Information: I am grateful to Xiaoou Li for discussing her work from (Chen et al. 2020), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of AmericaIIS-1837992 and BSF award2018230. Publisher Copyright: {\textcopyright} 2021 Elsevier B.V.",

year = "2021",

month = nov,

doi = "10.1016/j.spl.2021.109190",

language = "English (US)",

volume = "178",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

}

TY - JOUR

T1 - A note on identifiability conditions in confirmatory factor analysis

AU - Leeb, William

N1 - Funding Information: I am grateful to Xiaoou Li for discussing her work from ( Chen et al., 2020 ), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of America IIS-1837992 and BSF award 2018230 . Funding Information: I am grateful to Xiaoou Li for discussing her work from (Chen et al. 2020), and to the reviewers for their helpful comments. I acknowledge support from NSF award, United States of AmericaIIS-1837992 and BSF award2018230. Publisher Copyright: © 2021 Elsevier B.V.

PY - 2021/11

Y1 - 2021/11

N2 - Recently, Chen, Li and Zhang established conditions characterizing asymptotic identifiability of latent factors in confirmatory factor analysis. We give an elementary proof showing that a similar characterization holds non-asymptotically, and prove a related result for identifiability of factor loadings.

AB - Recently, Chen, Li and Zhang established conditions characterizing asymptotic identifiability of latent factors in confirmatory factor analysis. We give an elementary proof showing that a similar characterization holds non-asymptotically, and prove a related result for identifiability of factor loadings.

KW - Confirmatory factor analysis

KW - Factor loadings

KW - Identifiability

KW - Latent factors

KW - Matrix factorizations

UR - http://www.scopus.com/inward/record.url?scp=85111058256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85111058256&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2021.109190

DO - 10.1016/j.spl.2021.109190

M3 - Article

AN - SCOPUS:85111058256

SN - 0167-7152

VL - 178

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

M1 - 109190

ER -

A note on identifiability conditions in confirmatory factor analysis

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this