Home > Research > Publications & Outputs > Impact of an equality constraint on the class-s...

Electronic data

  • Revised_BR-Org-14-155

    Rights statement: The final publication is available at Springer via http://dx.doi.org/10.3758/s13428-015-0618-8

    Accepted author manuscript, 3.55 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Impact of an equality constraint on the class-specific residual variances in regression mixtures: a Monte Carlo simulation study

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Impact of an equality constraint on the class-specific residual variances in regression mixtures : a Monte Carlo simulation study. / Kim, Minjung; Lamont, Andrea E.; Jaki, Thomas et al.

In: Behavior Research Methods, Vol. 48, No. 2, 06.2016, p. 813-826.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Kim, M, Lamont, AE, Jaki, T, Feaster, D, Howe, G & Van Horn, ML 2016, 'Impact of an equality constraint on the class-specific residual variances in regression mixtures: a Monte Carlo simulation study', Behavior Research Methods, vol. 48, no. 2, pp. 813-826. https://doi.org/10.3758/s13428-015-0618-8

APA

Vancouver

Kim M, Lamont AE, Jaki T, Feaster D, Howe G, Van Horn ML. Impact of an equality constraint on the class-specific residual variances in regression mixtures: a Monte Carlo simulation study. Behavior Research Methods. 2016 Jun;48(2):813-826. Epub 2015 Jul 3. doi: 10.3758/s13428-015-0618-8

Author

Kim, Minjung ; Lamont, Andrea E. ; Jaki, Thomas et al. / Impact of an equality constraint on the class-specific residual variances in regression mixtures : a Monte Carlo simulation study. In: Behavior Research Methods. 2016 ; Vol. 48, No. 2. pp. 813-826.

Bibtex

@article{e7a44460e6514c2c82cb362a822ac312,
title = "Impact of an equality constraint on the class-specific residual variances in regression mixtures: a Monte Carlo simulation study",
abstract = "Regression mixture models are a novel approach to modeling the heterogeneous effects of predictors on an outcome. In the model-building process, often residual variances are disregarded and simplifying assumptions are made without thorough examination of the consequences. In this simulation study, we investigated the impact of an equality constraint on the residual variances across latent classes. We examined the consequences of constraining the residual variances on class enumeration (finding the true number of latent classes) and on the parameter estimates, under a number of different simulation conditions meant to reflect the types of heterogeneity likely to exist in applied analyses. The results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted on the estimated class sizes and showed the potential to greatly affect the parameter estimates in each class. These results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions are made.",
keywords = "Regression mixture, Differential effects, Effect heterogeneity, Residual variances",
author = "Minjung Kim and Lamont, {Andrea E.} and Thomas Jaki and Daniel Feaster and George Howe and {Van Horn}, {M. Lee}",
note = "The final publication is available at Springer via http://dx.doi.org/10.3758/s13428-015-0618-8",
year = "2016",
month = jun,
doi = "10.3758/s13428-015-0618-8",
language = "English",
volume = "48",
pages = "813--826",
journal = "Behavior Research Methods",
issn = "1554-351X",
publisher = "Springer New York LLC",
number = "2",

}

RIS

TY - JOUR

T1 - Impact of an equality constraint on the class-specific residual variances in regression mixtures

T2 - a Monte Carlo simulation study

AU - Kim, Minjung

AU - Lamont, Andrea E.

AU - Jaki, Thomas

AU - Feaster, Daniel

AU - Howe, George

AU - Van Horn, M. Lee

N1 - The final publication is available at Springer via http://dx.doi.org/10.3758/s13428-015-0618-8

PY - 2016/6

Y1 - 2016/6

N2 - Regression mixture models are a novel approach to modeling the heterogeneous effects of predictors on an outcome. In the model-building process, often residual variances are disregarded and simplifying assumptions are made without thorough examination of the consequences. In this simulation study, we investigated the impact of an equality constraint on the residual variances across latent classes. We examined the consequences of constraining the residual variances on class enumeration (finding the true number of latent classes) and on the parameter estimates, under a number of different simulation conditions meant to reflect the types of heterogeneity likely to exist in applied analyses. The results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted on the estimated class sizes and showed the potential to greatly affect the parameter estimates in each class. These results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions are made.

AB - Regression mixture models are a novel approach to modeling the heterogeneous effects of predictors on an outcome. In the model-building process, often residual variances are disregarded and simplifying assumptions are made without thorough examination of the consequences. In this simulation study, we investigated the impact of an equality constraint on the residual variances across latent classes. We examined the consequences of constraining the residual variances on class enumeration (finding the true number of latent classes) and on the parameter estimates, under a number of different simulation conditions meant to reflect the types of heterogeneity likely to exist in applied analyses. The results showed that bias in class enumeration increased as the difference in residual variances between the classes increased. Also, an inappropriate equality constraint on the residual variances greatly impacted on the estimated class sizes and showed the potential to greatly affect the parameter estimates in each class. These results suggest that it is important to make assumptions about residual variances with care and to carefully report what assumptions are made.

KW - Regression mixture

KW - Differential effects

KW - Effect heterogeneity

KW - Residual variances

U2 - 10.3758/s13428-015-0618-8

DO - 10.3758/s13428-015-0618-8

M3 - Journal article

VL - 48

SP - 813

EP - 826

JO - Behavior Research Methods

JF - Behavior Research Methods

SN - 1554-351X

IS - 2

ER -