A bias-adjusted estimator in quantile regression for clustered data
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Standard
A bias-adjusted estimator in quantile regression for clustered data. / Battagliola, Maria Laura; Sørensen, Helle; Tolver, Anders; Staicu, Ana Maria.
I: Econometrics and Statistics, Bind 23, 2022, s. 165-186.Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - A bias-adjusted estimator in quantile regression for clustered data
AU - Battagliola, Maria Laura
AU - Sørensen, Helle
AU - Tolver, Anders
AU - Staicu, Ana Maria
N1 - Publisher Copyright: © 2021 EcoSta Econometrics and Statistics
PY - 2022
Y1 - 2022
N2 - Quantile regression models with random effects are useful for studying associations between covariates and quantiles of the response distribution for clustered data. Parameter estimation is examined for a class of mixed-effects quantile regression models, with focus on settings with many but small clusters. The main contributions are the following: (i) documenting that existing methods may lead to severely biased estimators for fixed effects parameters; (ii) proposing a new two-step estimation methodology where predictions of the random effects are first computed by a pseudo likelihood approach (the LQMM method) and then used as offsets in standard quantile regression; (iii) proposing a novel bootstrap sampling procedure in order to reduce bias of the two-step estimator and compute confidence intervals. The proposed estimation and associated inference is assessed numerically through rigorous simulation studies and applied to an AIDS Clinical Trial Group (ACTG) study.
AB - Quantile regression models with random effects are useful for studying associations between covariates and quantiles of the response distribution for clustered data. Parameter estimation is examined for a class of mixed-effects quantile regression models, with focus on settings with many but small clusters. The main contributions are the following: (i) documenting that existing methods may lead to severely biased estimators for fixed effects parameters; (ii) proposing a new two-step estimation methodology where predictions of the random effects are first computed by a pseudo likelihood approach (the LQMM method) and then used as offsets in standard quantile regression; (iii) proposing a novel bootstrap sampling procedure in order to reduce bias of the two-step estimator and compute confidence intervals. The proposed estimation and associated inference is assessed numerically through rigorous simulation studies and applied to an AIDS Clinical Trial Group (ACTG) study.
KW - AIDS clinical trial group study
KW - Bias-adjustment
KW - Clustered data
KW - Linear quantile regression
KW - Random effects
KW - Wild bootstrap
UR - http://www.scopus.com/inward/record.url?scp=85120695107&partnerID=8YFLogxK
U2 - 10.1016/j.ecosta.2021.07.003
DO - 10.1016/j.ecosta.2021.07.003
M3 - Journal article
AN - SCOPUS:85120695107
VL - 23
SP - 165
EP - 186
JO - Econometrics and Statistics
JF - Econometrics and Statistics
SN - 2452-3062
ER -
ID: 291755082