Robust interactive fixed effects

Kris Boudt; Ewoud Heyndels

doi:10.1016/j.ecosta.2022.01.002

Robust interactive fixed effects

Kris Boudt^*, Ewoud Heyndels

^*Corresponding author for this work

Econometrics and Data Science

Research output: Contribution to Journal › Article › Academic › peer-review

6 Downloads (Pure)

Abstract

Robust estimators are proposed for the interactive fixed effects panel data model. In each iteration of the estimation algorithm the coefficients of the observable variables are estimated with robust regressions and the latent factors are extracted with robust principal component analysis. The reliability of the proposed procedure is documented in an extensive simulation study. The procedure is applied to cluster annual income growth time series of Belgian independents.

Original language	English
Pages (from-to)	206-223
Number of pages	18
Journal	Econometrics and Statistics
Volume	29
Early online date	21 Jan 2022
DOIs	https://doi.org/10.1016/j.ecosta.2022.01.002
Publication status	Published - Jan 2024

Bibliographical note

Funding Information:
We thank Innoviris for their funding and RSVZ for sharing their data and expertise. We are particularly grateful to Peter Arryn, Annick De Groot and Veerle De Maesschalck, as well as to Christophe Croux, Geert Dhaene, Guanglin Huang and Ruben Schoonacker for their feedback on earlier versions of the paper. We also thank Ricardo Maronna for sharing his code on pertMM, the authors of Alonso et al. (2020) for sharing theirs, and participants of the 2019 CFE workshop. We also thank the editor (Ana Colubi), the associate editor and three referees for their helpful comments.

Publisher Copyright:
© 2022 EcoSta Econometrics and Statistics

Keywords

clustering
factor structure
outlier
principal component
robust estimation
time series

Access to Document

10.1016/j.ecosta.2022.01.002

Robust interactive fixed effectsFinal published version, 1.8 MBLicence: Article 25fa Dutch Copyright Act

Persistent URL (handle)

Persistent link

Cite this

@article{1b710cd634ae42d2a2d854e699ca18b1,

title = "Robust interactive fixed effects",

abstract = "Robust estimators are proposed for the interactive fixed effects panel data model. In each iteration of the estimation algorithm the coefficients of the observable variables are estimated with robust regressions and the latent factors are extracted with robust principal component analysis. The reliability of the proposed procedure is documented in an extensive simulation study. The procedure is applied to cluster annual income growth time series of Belgian independents.",

keywords = "clustering, factor structure, outlier, principal component, robust estimation, time series",

author = "Kris Boudt and Ewoud Heyndels",

note = "Funding Information: We thank Innoviris for their funding and RSVZ for sharing their data and expertise. We are particularly grateful to Peter Arryn, Annick De Groot and Veerle De Maesschalck, as well as to Christophe Croux, Geert Dhaene, Guanglin Huang and Ruben Schoonacker for their feedback on earlier versions of the paper. We also thank Ricardo Maronna for sharing his code on pertMM, the authors of Alonso et al. (2020) for sharing theirs, and participants of the 2019 CFE workshop. We also thank the editor (Ana Colubi), the associate editor and three referees for their helpful comments. Publisher Copyright: {\textcopyright} 2022 EcoSta Econometrics and Statistics",

year = "2024",

month = jan,

doi = "10.1016/j.ecosta.2022.01.002",

language = "English",

volume = "29",

pages = "206--223",

journal = "Econometrics and Statistics",

issn = "2452-3062",

publisher = "Elsevier BV",

}

TY - JOUR

T1 - Robust interactive fixed effects

AU - Boudt, Kris

AU - Heyndels, Ewoud

N1 - Funding Information: We thank Innoviris for their funding and RSVZ for sharing their data and expertise. We are particularly grateful to Peter Arryn, Annick De Groot and Veerle De Maesschalck, as well as to Christophe Croux, Geert Dhaene, Guanglin Huang and Ruben Schoonacker for their feedback on earlier versions of the paper. We also thank Ricardo Maronna for sharing his code on pertMM, the authors of Alonso et al. (2020) for sharing theirs, and participants of the 2019 CFE workshop. We also thank the editor (Ana Colubi), the associate editor and three referees for their helpful comments. Publisher Copyright: © 2022 EcoSta Econometrics and Statistics

PY - 2024/1

Y1 - 2024/1

N2 - Robust estimators are proposed for the interactive fixed effects panel data model. In each iteration of the estimation algorithm the coefficients of the observable variables are estimated with robust regressions and the latent factors are extracted with robust principal component analysis. The reliability of the proposed procedure is documented in an extensive simulation study. The procedure is applied to cluster annual income growth time series of Belgian independents.

AB - Robust estimators are proposed for the interactive fixed effects panel data model. In each iteration of the estimation algorithm the coefficients of the observable variables are estimated with robust regressions and the latent factors are extracted with robust principal component analysis. The reliability of the proposed procedure is documented in an extensive simulation study. The procedure is applied to cluster annual income growth time series of Belgian independents.

KW - clustering

KW - factor structure

KW - outlier

KW - principal component

KW - robust estimation

KW - time series

UR - http://www.scopus.com/inward/record.url?scp=85124032906&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85124032906&partnerID=8YFLogxK

U2 - 10.1016/j.ecosta.2022.01.002

DO - 10.1016/j.ecosta.2022.01.002

M3 - Article

AN - SCOPUS:85124032906

SN - 2452-3062

VL - 29

SP - 206

EP - 223

JO - Econometrics and Statistics

JF - Econometrics and Statistics

ER -

Robust interactive fixed effects

Abstract

Bibliographical note

Keywords

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this