A Diagnostic Procedure for Detecting Outliers in Linear State–Space Models

Dongjun You, Michael Hunter, Meng Chen, Sy-Miin Chow

Research output: Contribution to journalArticle

Abstract

Outliers can be more problematic in longitudinal data than in independent observations due to the correlated nature of such data. It is common practice to discard outliers as they are typically regarded as a nuisance or an aberration in the data. However, outliers can also convey meaningful information concerning potential model misspecification, and ways to modify and improve the model. Moreover, outliers that occur among the latent variables (innovative outliers) have distinct characteristics compared to those impacting the observed variables (additive outliers), and are best evaluated with different test statistics and detection procedures. We demonstrate and evaluate the performance of an outlier detection approach for multi-subject state-space models in a Monte Carlo simulation study, with corresponding adaptations to improve power and reduce false detection rates. Furthermore, we demonstrate the empirical utility of the proposed approach using data from an ecological momentary assessment study of emotion regulation together with an open-source software implementation of the procedures.

Original languageEnglish (US)
JournalMultivariate Behavioral Research
DOIs
StatePublished - Jan 1 2019

Fingerprint

State-space Model
Outlier
Linear Models
Linear Model
Diagnostics
Space Simulation
Emotions
Software
Additive Outliers
Model Misspecification
Multi-state
Open Source Software
Outlier Detection
Latent Variables
Longitudinal Data
Aberration
Demonstrate
Test Statistic
Monte Carlo Simulation
Outliers

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Experimental and Cognitive Psychology
  • Arts and Humanities (miscellaneous)

Cite this

@article{96baf4745c4b4ae3a464e471186a6ce6,
title = "A Diagnostic Procedure for Detecting Outliers in Linear State–Space Models",
abstract = "Outliers can be more problematic in longitudinal data than in independent observations due to the correlated nature of such data. It is common practice to discard outliers as they are typically regarded as a nuisance or an aberration in the data. However, outliers can also convey meaningful information concerning potential model misspecification, and ways to modify and improve the model. Moreover, outliers that occur among the latent variables (innovative outliers) have distinct characteristics compared to those impacting the observed variables (additive outliers), and are best evaluated with different test statistics and detection procedures. We demonstrate and evaluate the performance of an outlier detection approach for multi-subject state-space models in a Monte Carlo simulation study, with corresponding adaptations to improve power and reduce false detection rates. Furthermore, we demonstrate the empirical utility of the proposed approach using data from an ecological momentary assessment study of emotion regulation together with an open-source software implementation of the procedures.",
author = "Dongjun You and Michael Hunter and Meng Chen and Sy-Miin Chow",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/00273171.2019.1627659",
language = "English (US)",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",

}

A Diagnostic Procedure for Detecting Outliers in Linear State–Space Models. / You, Dongjun; Hunter, Michael; Chen, Meng; Chow, Sy-Miin.

In: Multivariate Behavioral Research, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Diagnostic Procedure for Detecting Outliers in Linear State–Space Models

AU - You, Dongjun

AU - Hunter, Michael

AU - Chen, Meng

AU - Chow, Sy-Miin

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Outliers can be more problematic in longitudinal data than in independent observations due to the correlated nature of such data. It is common practice to discard outliers as they are typically regarded as a nuisance or an aberration in the data. However, outliers can also convey meaningful information concerning potential model misspecification, and ways to modify and improve the model. Moreover, outliers that occur among the latent variables (innovative outliers) have distinct characteristics compared to those impacting the observed variables (additive outliers), and are best evaluated with different test statistics and detection procedures. We demonstrate and evaluate the performance of an outlier detection approach for multi-subject state-space models in a Monte Carlo simulation study, with corresponding adaptations to improve power and reduce false detection rates. Furthermore, we demonstrate the empirical utility of the proposed approach using data from an ecological momentary assessment study of emotion regulation together with an open-source software implementation of the procedures.

AB - Outliers can be more problematic in longitudinal data than in independent observations due to the correlated nature of such data. It is common practice to discard outliers as they are typically regarded as a nuisance or an aberration in the data. However, outliers can also convey meaningful information concerning potential model misspecification, and ways to modify and improve the model. Moreover, outliers that occur among the latent variables (innovative outliers) have distinct characteristics compared to those impacting the observed variables (additive outliers), and are best evaluated with different test statistics and detection procedures. We demonstrate and evaluate the performance of an outlier detection approach for multi-subject state-space models in a Monte Carlo simulation study, with corresponding adaptations to improve power and reduce false detection rates. Furthermore, we demonstrate the empirical utility of the proposed approach using data from an ecological momentary assessment study of emotion regulation together with an open-source software implementation of the procedures.

UR - http://www.scopus.com/inward/record.url?scp=85068554496&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068554496&partnerID=8YFLogxK

U2 - 10.1080/00273171.2019.1627659

DO - 10.1080/00273171.2019.1627659

M3 - Article

C2 - 31264463

AN - SCOPUS:85068554496

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

ER -