A Regularized GLS for Structural Equation Modeling
Authors:
- 16 May 2017
- journal article
- research article
- Published by Taylor & Francis in Structural Equation Modeling: A Multidisciplinary Journal
- Vol. 24 (5) , 657-665
Abstract
Ill conditioning of covariance and weight matrices used in structural equation modeling (SEM) is a possible source of inadequate performance of SEM statistics in nonasymptotic samples. A maximum a posteriori (MAP) covariance matrix is proposed for weight matrix regularization in normal theory generalized least squares (GLS) estimation. Maximum likelihood (ML), GLS, and regularized GLS test statistics (RGLS and rGLS) are studied by simulation in a 15-variable, 3-factor model with 15 levels of sample size varying from 60 to 100,000. A key result showed that in terms of nominal rejection rates, RGLS outperformed ML at all sample sizes below 500, and GLS at most sample sizes below 500. In larger samples, their performance was equivalent. The second regularization methodology (rGLS) performed well asymptotically, but poorly in small samples. Regularization in SEM deserves further study.Keywords
This publication has 37 references indexed in Scilit:
- The Model Size Effect in SEM: Inflated Goodness-of-Fit Statistics Are Due to the Size of the Covariance MatrixStructural Equation Modeling: A Multidisciplinary Journal, 2012
- Accuracy in parameter estimation for targeted effects in structural equation modeling: Sample size planning for narrow confidence intervals.Psychological Methods, 2011
- The Model-Size Effect on Traditional and Modified Tests of Covariance StructuresStructural Equation Modeling: A Multidisciplinary Journal, 2007
- The use of item parcels in structural equation modelling: Non‐normal data and small sample sizesBritish Journal of Mathematical and Statistical Psychology, 2004
- A well-conditioned estimator for large-dimensional covariance matricesJournal of Multivariate Analysis, 2004
- Performance of Modified Test Statistics in Covariance and Correlation Structure Analysis Under Conditions of Multivariate NonnormalityStructural Equation Modeling: A Multidisciplinary Journal, 2000
- Can test statistics in covariance structure analysis be trusted?Psychological Bulletin, 1992
- A note on computational efficiency in asymptotically distribution‐free correlational modelsBritish Journal of Mathematical and Statistical Psychology, 1985
- COVARIANCE STRUCTURES1982
- TESTS OF SIGNIFICANCE IN FACTOR ANALYSISBritish Journal of Statistical Psychology, 1950