Regularized Structural Equation Modeling

Ross Jacobucci, Kevin Grimm, John J. McArdle

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

A new method is proposed that extends the use of regularization in both lasso and ridge regression to structural equation models. The method is termed regularized structural equation modeling (RegSEM). RegSEM penalizes specific parameters in structural equation models, with the goal of creating easier to understand and simpler models. Although regularization has gained wide adoption in regression, very little has transferred to models with latent variables. By adding penalties to specific parameters in a structural equation model, researchers have a high level of flexibility in reducing model complexity, overcoming poor fitting models, and the creation of models that are more likely to generalize to new samples. The proposed method was evaluated through a simulation study, two illustrative examples involving a measurement model, and one empirical example involving the structural part of the model to demonstrate RegSEM’s utility.

Original languageEnglish (US)
Pages (from-to)1-12
Number of pages12
JournalStructural Equation Modeling
DOIs
StateAccepted/In press - Apr 13 2016

Fingerprint

Structural Equation Modeling
Structural Equation Model
structural model
Regularization
Ridge Regression
Lasso
Model Complexity
Model
Model Fitting
Latent Variables
regression
Penalty
Regression
Likely
Flexibility
Structural equation modeling
Simulation Study
penalty
Generalise
flexibility

Keywords

  • factor analysis
  • lasso
  • penalization
  • regularization
  • ridge
  • shrinkage
  • structural equation modeling

ASJC Scopus subject areas

  • Modeling and Simulation
  • Decision Sciences(all)
  • Economics, Econometrics and Finance(all)
  • Sociology and Political Science

Cite this

Regularized Structural Equation Modeling. / Jacobucci, Ross; Grimm, Kevin; McArdle, John J.

In: Structural Equation Modeling, 13.04.2016, p. 1-12.

Research output: Contribution to journalArticle

Jacobucci, Ross ; Grimm, Kevin ; McArdle, John J. / Regularized Structural Equation Modeling. In: Structural Equation Modeling. 2016 ; pp. 1-12.
@article{68cfe379349c4098b93dee86170f0e8d,
title = "Regularized Structural Equation Modeling",
abstract = "A new method is proposed that extends the use of regularization in both lasso and ridge regression to structural equation models. The method is termed regularized structural equation modeling (RegSEM). RegSEM penalizes specific parameters in structural equation models, with the goal of creating easier to understand and simpler models. Although regularization has gained wide adoption in regression, very little has transferred to models with latent variables. By adding penalties to specific parameters in a structural equation model, researchers have a high level of flexibility in reducing model complexity, overcoming poor fitting models, and the creation of models that are more likely to generalize to new samples. The proposed method was evaluated through a simulation study, two illustrative examples involving a measurement model, and one empirical example involving the structural part of the model to demonstrate RegSEM’s utility.",
keywords = "factor analysis, lasso, penalization, regularization, ridge, shrinkage, structural equation modeling",
author = "Ross Jacobucci and Kevin Grimm and McArdle, {John J.}",
year = "2016",
month = "4",
day = "13",
doi = "10.1080/10705511.2016.1154793",
language = "English (US)",
pages = "1--12",
journal = "Structural Equation Modeling",
issn = "1070-5511",
publisher = "Psychology Press Ltd",

}

TY - JOUR

T1 - Regularized Structural Equation Modeling

AU - Jacobucci, Ross

AU - Grimm, Kevin

AU - McArdle, John J.

PY - 2016/4/13

Y1 - 2016/4/13

N2 - A new method is proposed that extends the use of regularization in both lasso and ridge regression to structural equation models. The method is termed regularized structural equation modeling (RegSEM). RegSEM penalizes specific parameters in structural equation models, with the goal of creating easier to understand and simpler models. Although regularization has gained wide adoption in regression, very little has transferred to models with latent variables. By adding penalties to specific parameters in a structural equation model, researchers have a high level of flexibility in reducing model complexity, overcoming poor fitting models, and the creation of models that are more likely to generalize to new samples. The proposed method was evaluated through a simulation study, two illustrative examples involving a measurement model, and one empirical example involving the structural part of the model to demonstrate RegSEM’s utility.

AB - A new method is proposed that extends the use of regularization in both lasso and ridge regression to structural equation models. The method is termed regularized structural equation modeling (RegSEM). RegSEM penalizes specific parameters in structural equation models, with the goal of creating easier to understand and simpler models. Although regularization has gained wide adoption in regression, very little has transferred to models with latent variables. By adding penalties to specific parameters in a structural equation model, researchers have a high level of flexibility in reducing model complexity, overcoming poor fitting models, and the creation of models that are more likely to generalize to new samples. The proposed method was evaluated through a simulation study, two illustrative examples involving a measurement model, and one empirical example involving the structural part of the model to demonstrate RegSEM’s utility.

KW - factor analysis

KW - lasso

KW - penalization

KW - regularization

KW - ridge

KW - shrinkage

KW - structural equation modeling

UR - http://www.scopus.com/inward/record.url?scp=84963576551&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963576551&partnerID=8YFLogxK

U2 - 10.1080/10705511.2016.1154793

DO - 10.1080/10705511.2016.1154793

M3 - Article

AN - SCOPUS:84963576551

SP - 1

EP - 12

JO - Structural Equation Modeling

JF - Structural Equation Modeling

SN - 1070-5511

ER -