Implementing Restricted Maximum Likelihood Estimation in Structural Equation Models

By Mike W.-L. Cheung. 2013

54 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Management, Monitoring, Policy and Law, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/e2f63e9f0c20601252bcaa68071ea8f985bebfa1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

1. Daniel M. McNeish
2. Laura M. Stapleton
3. Mike W.-L. Cheung
4. Daniel L. Oberski
5. T. Asparouhov
6. Mei-Shiu Chiu
7. Kristopher J Preacher
8. Matthew A. Diemer

In Management, Monitoring, Policy and Law

1. Daniel M. McNeish
2. Laura M. Stapleton
3. Mike W.-L. Cheung

In Management Science and Operations Research

1. Daniel M. McNeish
2. Laura M. Stapleton
3. Mike W.-L. Cheung
4. Daniel L. Oberski
5. Andrew F. Hayes
6. T. Asparouhov
7. Mei-Shiu Chiu
8. Kristopher J Preacher

Citations in the corpus

• metaSEM: an R package for meta-analysis using structural equation modeling (2015, methodology): “It formulates univariate, multivariate, and three-level meta-analytic models as structural equation models (Cheung, 2008, 2013b, 2014c, in press) via the OpenMx package (Boker et al., 2011).”
• Modeling dependent effect sizes with three-level meta-analyses: a structural equation modeling approach. (2014, methodology): “…multivariate meta-analyses (M. W.-L. Cheung, 2008, in press), whereas reml() implements the same analysis using the REML estimation method (M. W.-L. Cheung, 2013). tssem1() and tssem2() conduct meta-analytic structural equation modeling using the two-stage SEM approach (M. W.-L. Cheung & Chan,…”
• Multivariate Meta-Analysis as Structural Equation Models (2013, methodology): “To partially address this concern, a free R package called metaSEM (Cheung, 2013b) implemented in OpenMx and R (R Development Core Team, 2013) has been written.”
• Multilevel Mediation With Small Samples: A Cautionary Note on the Multilevel Structural Equation Modeling Framework (2017, methodology): “Although some work has advanced rudimentary forms of REML for the specific types of models that can be fit as SEMs (mostly growth models; Cheung, 2013), applications of REML for SEM are still extremely limited generally and no statistical theory exists for REML estimation of ML-SEM (for…”
• Modeling Clustered Data with Very Few Clusters (2016, background): “…the lme4 R package, HLM 7) is restricted maximum likelihood (REML), which is known to exhibit better finite sample properties compared to traditional maximum likelihood, especially for estimates of the elements of the G matrix (e.g., Browne & Draper, 2006; Cheung, 2013; McNeish & Stapleton, 2014).”