Advanced search
2 files | 2.60 MB Add to list

Multilevel autoregressive models when the number of time points is small

Fien Gistelinck, Tom Loeys (UGent) and Nele Flamant (UGent)
Author
Organization
Abstract
The multilevel autoregressive model disentangles unobserved heterogeneity from state-dependence. Statistically, the random intercept accounts for the dependence of all measurements at different time points on an observed underlying factor, while the lagged dependent predictor allows the outcome to depend on the outcome at the previous time point. In this paper, we consider different implementations of the simplest multilevel autoregressive model, and explore how each of them deals with the endogeneity assumption and the initial conditions problem. We discuss the performance of the no centering approach, the manifest centering approach, and the latent centering approach in the setting where the number of time points is small. We find that some commonly used approaches show bias for the autoregressive parameter. When the outcome at the first time point is considered predetermined, the no centering approach assuming endogeneity performs best.
Keywords
BAYESIAN-APPROACH, Structural equation modeling, latent centering, multilevel autoregressive models, panel data

Downloads

  • LDVs rev1.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 360.93 KB
  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 2.24 MB

Citation

Please use this url to cite or link to this publication:

MLA
Gistelinck, Fien, et al. “Multilevel Autoregressive Models When the Number of Time Points Is Small.” STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, vol. 28, no. 1, 2021, pp. 15–27, doi:10.1080/10705511.2020.1753517.
APA
Gistelinck, F., Loeys, T., & Flamant, N. (2021). Multilevel autoregressive models when the number of time points is small. STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, 28(1), 15–27. https://doi.org/10.1080/10705511.2020.1753517
Chicago author-date
Gistelinck, Fien, Tom Loeys, and Nele Flamant. 2021. “Multilevel Autoregressive Models When the Number of Time Points Is Small.” STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL 28 (1): 15–27. https://doi.org/10.1080/10705511.2020.1753517.
Chicago author-date (all authors)
Gistelinck, Fien, Tom Loeys, and Nele Flamant. 2021. “Multilevel Autoregressive Models When the Number of Time Points Is Small.” STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL 28 (1): 15–27. doi:10.1080/10705511.2020.1753517.
Vancouver
1.
Gistelinck F, Loeys T, Flamant N. Multilevel autoregressive models when the number of time points is small. STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL. 2021;28(1):15–27.
IEEE
[1]
F. Gistelinck, T. Loeys, and N. Flamant, “Multilevel autoregressive models when the number of time points is small,” STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL, vol. 28, no. 1, pp. 15–27, 2021.
@article{8695151,
  abstract     = {{The multilevel autoregressive model disentangles unobserved heterogeneity from state-dependence. Statistically, the random intercept accounts for the dependence of all measurements at different time points on an observed underlying factor, while the lagged dependent predictor allows the outcome to depend on the outcome at the previous time point. In this paper, we consider different implementations of the simplest multilevel autoregressive model, and explore how each of them deals with the endogeneity assumption and the initial conditions problem. We discuss the performance of the no centering approach, the manifest centering approach, and the latent centering approach in the setting where the number of time points is small. We find that some commonly used approaches show bias for the autoregressive parameter. When the outcome at the first time point is considered predetermined, the no centering approach assuming endogeneity performs best.}},
  author       = {{Gistelinck, Fien and Loeys, Tom and Flamant, Nele}},
  issn         = {{1070-5511}},
  journal      = {{STRUCTURAL EQUATION MODELING-A MULTIDISCIPLINARY JOURNAL}},
  keywords     = {{BAYESIAN-APPROACH,Structural equation modeling,latent centering,multilevel autoregressive models,panel data}},
  language     = {{eng}},
  number       = {{1}},
  pages        = {{15--27}},
  title        = {{Multilevel autoregressive models when the number of time points is small}},
  url          = {{http://doi.org/10.1080/10705511.2020.1753517}},
  volume       = {{28}},
  year         = {{2021}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: