A Bayesian Solution to Non-convergence of Crossed Random Effects Models

Mingya Huang, Carolyn Anderson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Crossed random effects models can simultaneously take into account both fixed effects and random effects of the subjects and stimuli when observations are nested within combinations of subjects and stimuli. Unfortunately, maximum likelihood estimation (MLE) and restricted maximum likelihood (REML) estimation often encounter convergence problems, which in turn lead researchers to fit simpler models that yield invalid statistical inferences. On the other hand, if the random effects structure is too simple, tests of fixed effects are not valid; if the random effect structure is too complex, tests of fixed effects are inefficient. This study examines non-convergence issues inherent with MLE and REML as well as whether using Bayesian estimation can solve estimation problems when using crossed random effects models.

Original languageEnglish (US)
Title of host publicationQuantitative Psychology - The 85th Annual Meeting of the Psychometric Society
EditorsMarie Wiberg, Dylan Molenaar, Jorge González, Ulf Böckenholt, Jee-Seon Kim
PublisherSpringer
Pages297-307
Number of pages11
ISBN (Print)9783030747718
DOIs
StatePublished - 2021
Event85th Annual International Meeting of the Psychometric Society, IMPS 2020 - Virtual, Online
Duration: Jul 13 2020Jul 17 2020

Publication series

NameSpringer Proceedings in Mathematics and Statistics
Volume353
ISSN (Print)2194-1009
ISSN (Electronic)2194-1017

Conference

Conference85th Annual International Meeting of the Psychometric Society, IMPS 2020
CityVirtual, Online
Period7/13/207/17/20

Keywords

  • Bayesian estimation
  • Convergence
  • Crossed-random effects models
  • Psycholinguistics

ASJC Scopus subject areas

  • General Mathematics

Fingerprint

Dive into the research topics of 'A Bayesian Solution to Non-convergence of Crossed Random Effects Models'. Together they form a unique fingerprint.

Cite this