Model choice and diagnostics for linear mixed-effects models using statistics on street corners

Adam Loy, Heike Hofmann, Dianne Cook

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The complexity of linear mixed-effects (LME) models means that traditional diagnostics are rendered less effective. This is due to a breakdown of asymptotic results, boundary issues, and visible patterns in residual plots that are introduced by the model fitting process. Some of these issues are well known and adjustments have been proposed. Working with LME models typically requires that the analyst keeps track of all the special circumstances that may arise. In this article, we illustrate a simpler but generally applicable approach to diagnosing LME models. We explain how to use new visual inference methods for these purposes. The approach provides a unified framework for diagnosing LME fits and for model selection. We illustrate the use of this approach on several commonly available datasets. A large-scale Amazon Turk study was used to validate the methods. R code is provided for the analyses. Supplementary materials for this article are available online.

LanguageEnglish
Pages478-492
Number of pages15
JournalJournal of Computational and Graphical Statistics
Volume26
Issue number3
DOIs
Publication statusPublished - 2017

Keywords

  • Lineup protocol
  • Model diagnostics
  • Model selection
  • Statistical graphics
  • Visual inference

Cite this

@article{918ebbc29aa7401f84f479699b37c419,
title = "Model choice and diagnostics for linear mixed-effects models using statistics on street corners",
abstract = "The complexity of linear mixed-effects (LME) models means that traditional diagnostics are rendered less effective. This is due to a breakdown of asymptotic results, boundary issues, and visible patterns in residual plots that are introduced by the model fitting process. Some of these issues are well known and adjustments have been proposed. Working with LME models typically requires that the analyst keeps track of all the special circumstances that may arise. In this article, we illustrate a simpler but generally applicable approach to diagnosing LME models. We explain how to use new visual inference methods for these purposes. The approach provides a unified framework for diagnosing LME fits and for model selection. We illustrate the use of this approach on several commonly available datasets. A large-scale Amazon Turk study was used to validate the methods. R code is provided for the analyses. Supplementary materials for this article are available online.",
keywords = "Lineup protocol, Model diagnostics, Model selection, Statistical graphics, Visual inference",
author = "Adam Loy and Heike Hofmann and Dianne Cook",
year = "2017",
doi = "10.1080/10618600.2017.1330207",
language = "English",
volume = "26",
pages = "478--492",
journal = "Journal of Computational and Graphical Statistics",
issn = "1061-8600",
publisher = "Taylor & Francis",
number = "3",

}

Model choice and diagnostics for linear mixed-effects models using statistics on street corners. / Loy, Adam; Hofmann, Heike; Cook, Dianne.

In: Journal of Computational and Graphical Statistics, Vol. 26, No. 3, 2017, p. 478-492.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Model choice and diagnostics for linear mixed-effects models using statistics on street corners

AU - Loy, Adam

AU - Hofmann, Heike

AU - Cook, Dianne

PY - 2017

Y1 - 2017

N2 - The complexity of linear mixed-effects (LME) models means that traditional diagnostics are rendered less effective. This is due to a breakdown of asymptotic results, boundary issues, and visible patterns in residual plots that are introduced by the model fitting process. Some of these issues are well known and adjustments have been proposed. Working with LME models typically requires that the analyst keeps track of all the special circumstances that may arise. In this article, we illustrate a simpler but generally applicable approach to diagnosing LME models. We explain how to use new visual inference methods for these purposes. The approach provides a unified framework for diagnosing LME fits and for model selection. We illustrate the use of this approach on several commonly available datasets. A large-scale Amazon Turk study was used to validate the methods. R code is provided for the analyses. Supplementary materials for this article are available online.

AB - The complexity of linear mixed-effects (LME) models means that traditional diagnostics are rendered less effective. This is due to a breakdown of asymptotic results, boundary issues, and visible patterns in residual plots that are introduced by the model fitting process. Some of these issues are well known and adjustments have been proposed. Working with LME models typically requires that the analyst keeps track of all the special circumstances that may arise. In this article, we illustrate a simpler but generally applicable approach to diagnosing LME models. We explain how to use new visual inference methods for these purposes. The approach provides a unified framework for diagnosing LME fits and for model selection. We illustrate the use of this approach on several commonly available datasets. A large-scale Amazon Turk study was used to validate the methods. R code is provided for the analyses. Supplementary materials for this article are available online.

KW - Lineup protocol

KW - Model diagnostics

KW - Model selection

KW - Statistical graphics

KW - Visual inference

UR - http://www.scopus.com/inward/record.url?scp=85026418622&partnerID=8YFLogxK

U2 - 10.1080/10618600.2017.1330207

DO - 10.1080/10618600.2017.1330207

M3 - Article

VL - 26

SP - 478

EP - 492

JO - Journal of Computational and Graphical Statistics

T2 - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

SN - 1061-8600

IS - 3

ER -