Asymptotic comparison of missing data procedures for estimating factor loadings

C. Hendricks Brown

Research output: Contribution to journalArticle

47 Citations (Scopus)

Abstract

Large sample properties of four methods of handling multivariate missing data are compared. The criterion for comparison is how well the loadings from a single factor model can be estimated. It is shown that efficiencies of the methods depend on the pattern or arrangement of missing data, and an evaluation study is used to generate predictive efficiency equations to guide one's choice of an estimating procedure. A simple regression-type estimator is introduced which shows high efficiency relative to the maximum likelihood method over a large range of patterns and covariance matrices.

Original languageEnglish
Pages (from-to)269-291
Number of pages23
JournalPsychometrika
Volume48
Issue number2
DOIs
StatePublished - Jun 1 1983

Fingerprint

Missing Data
efficiency
Factor Models
Maximum Likelihood Method
Multivariate Data
Covariance matrix
High Efficiency
Arrangement
Regression
Estimator
Maximum likelihood
Evaluation
Range of data
regression
evaluation

Keywords

  • EM algorithm

ASJC Scopus subject areas

  • Social Sciences (miscellaneous)
  • Psychology (miscellaneous)
  • Psychology(all)
  • Mathematics (miscellaneous)

Cite this

Asymptotic comparison of missing data procedures for estimating factor loadings. / Hendricks Brown, C.

In: Psychometrika, Vol. 48, No. 2, 01.06.1983, p. 269-291.

Research output: Contribution to journalArticle

Hendricks Brown, C. / Asymptotic comparison of missing data procedures for estimating factor loadings. In: Psychometrika. 1983 ; Vol. 48, No. 2. pp. 269-291.
@article{7874297267204cdf8eb3f68cca8fa13e,
title = "Asymptotic comparison of missing data procedures for estimating factor loadings",
abstract = "Large sample properties of four methods of handling multivariate missing data are compared. The criterion for comparison is how well the loadings from a single factor model can be estimated. It is shown that efficiencies of the methods depend on the pattern or arrangement of missing data, and an evaluation study is used to generate predictive efficiency equations to guide one's choice of an estimating procedure. A simple regression-type estimator is introduced which shows high efficiency relative to the maximum likelihood method over a large range of patterns and covariance matrices.",
keywords = "EM algorithm",
author = "{Hendricks Brown}, C.",
year = "1983",
month = "6",
day = "1",
doi = "10.1007/BF02294022",
language = "English",
volume = "48",
pages = "269--291",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "2",

}

TY - JOUR

T1 - Asymptotic comparison of missing data procedures for estimating factor loadings

AU - Hendricks Brown, C.

PY - 1983/6/1

Y1 - 1983/6/1

N2 - Large sample properties of four methods of handling multivariate missing data are compared. The criterion for comparison is how well the loadings from a single factor model can be estimated. It is shown that efficiencies of the methods depend on the pattern or arrangement of missing data, and an evaluation study is used to generate predictive efficiency equations to guide one's choice of an estimating procedure. A simple regression-type estimator is introduced which shows high efficiency relative to the maximum likelihood method over a large range of patterns and covariance matrices.

AB - Large sample properties of four methods of handling multivariate missing data are compared. The criterion for comparison is how well the loadings from a single factor model can be estimated. It is shown that efficiencies of the methods depend on the pattern or arrangement of missing data, and an evaluation study is used to generate predictive efficiency equations to guide one's choice of an estimating procedure. A simple regression-type estimator is introduced which shows high efficiency relative to the maximum likelihood method over a large range of patterns and covariance matrices.

KW - EM algorithm

UR - http://www.scopus.com/inward/record.url?scp=0001326815&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001326815&partnerID=8YFLogxK

U2 - 10.1007/BF02294022

DO - 10.1007/BF02294022

M3 - Article

AN - SCOPUS:0001326815

VL - 48

SP - 269

EP - 291

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 2

ER -