Frequentist Model Averaging in Structural Equation Modelling.

Shaobo Jin, Sebastian Ankargren
Author Information
  1. Shaobo Jin: Department of Statistics, Uppsala University, Uppsala, Sweden. shaobo.jin@statistik.uu.se.
  2. Sebastian Ankargren: Department of Statistics, Uppsala University, Uppsala, Sweden.

Abstract

Model selection from a set of candidate models plays an important role in many structural equation modelling applications. However, traditional model selection methods introduce extra randomness that is not accounted for by post-model selection inference. In the current study, we propose a model averaging technique within the frequentist statistical framework. Instead of selecting an optimal model, the contributions of all candidate models are acknowledged. Valid confidence intervals and a [Formula: see text] test statistic are proposed. A simulation study shows that the proposed method is able to produce a robust mean-squared error, a better coverage probability, and a better goodness-of-fit test compared to model selection. It is an interesting compromise between model selection and the full model.

Keywords

References

  1. Psychol Bull. 1992 May;111(3):490-504 [PMID: 16250105]
  2. Psychol Bull. 1990 Mar;107(2):238-46 [PMID: 2320703]
  3. Br J Math Stat Psychol. 1984 May;37 ( Pt 1):62-83 [PMID: 6733054]

MeSH Term

Computer Simulation
Data Interpretation, Statistical
Humans
Intelligence
Interpersonal Relations
Latent Class Analysis
Learning
Models, Statistical
Psychometrics
Teaching
Universities

Word Cloud

Created with Highcharts 10.0.0modelselectionModelcandidatemodelsinferencestudytestproposedbettercoverageprobabilitygoodness-of-fitsetplaysimportantrolemanystructuralequationmodellingapplicationsHowevertraditionalmethodsintroduceextrarandomnessaccountedpost-modelcurrentproposeaveragingtechniquewithinfrequentiststatisticalframeworkInsteadselectingoptimalcontributionsacknowledgedValidconfidenceintervals[Formula:seetext]statisticsimulationshowsmethodableproducerobustmean-squarederrorcomparedinterestingcompromisefullFrequentistAveragingStructuralEquationModellinglocalasymptoticpost-selection

Similar Articles

Cited By