Bayesian partial linear model for skewed longitudinal data

Yuanyuan Tang, Debajyoti Sinha, Debdeep Pati, Stuart Lipsitz, Steven E Lipshultz

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Unlike majority of current statistical models and methods focusing on mean response for highly skewed longitudinal data, we present a novel model for such data accommodating a partially linear median regression function, a skewed error distribution and within subject association structures. We provide theoretical justifications for our methods including asymptotic properties of the posterior and associated semiparametric Bayesian estimators. We also provide simulation studies to investigate the finite sample properties of our methods. Several advantages of our method compared with existing methods are demonstrated via analysis of a cardiotoxicity study of children of HIV-infected mothers.

Original languageEnglish (US)
Pages (from-to)441-453
Number of pages13
JournalBiostatistics
Volume16
Issue number3
DOIs
StatePublished - Sep 17 2014
Externally publishedYes

Fingerprint

Partial Linear Model
Longitudinal Data
Linear Models
Median Regression
Bayesian Estimator
Regression Function
Justification
Statistical method
Statistical Model
Asymptotic Properties
Statistical Models
Simulation Study
Mothers
Partial linear model
Longitudinal data
HIV

Keywords

  • Dirichlet process
  • Median regression
  • Partial linear model
  • Semiparametric
  • Skewed error.

ASJC Scopus subject areas

  • Medicine(all)
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Tang, Y., Sinha, D., Pati, D., Lipsitz, S., & Lipshultz, S. E. (2014). Bayesian partial linear model for skewed longitudinal data. Biostatistics, 16(3), 441-453. https://doi.org/10.1093/biostatistics/kxv005

Bayesian partial linear model for skewed longitudinal data. / Tang, Yuanyuan; Sinha, Debajyoti; Pati, Debdeep; Lipsitz, Stuart; Lipshultz, Steven E.

In: Biostatistics, Vol. 16, No. 3, 17.09.2014, p. 441-453.

Research output: Contribution to journalArticle

Tang, Y, Sinha, D, Pati, D, Lipsitz, S & Lipshultz, SE 2014, 'Bayesian partial linear model for skewed longitudinal data', Biostatistics, vol. 16, no. 3, pp. 441-453. https://doi.org/10.1093/biostatistics/kxv005
Tang Y, Sinha D, Pati D, Lipsitz S, Lipshultz SE. Bayesian partial linear model for skewed longitudinal data. Biostatistics. 2014 Sep 17;16(3):441-453. https://doi.org/10.1093/biostatistics/kxv005
Tang, Yuanyuan ; Sinha, Debajyoti ; Pati, Debdeep ; Lipsitz, Stuart ; Lipshultz, Steven E. / Bayesian partial linear model for skewed longitudinal data. In: Biostatistics. 2014 ; Vol. 16, No. 3. pp. 441-453.
@article{994cdbfcefe14c7297b909929deb664f,
title = "Bayesian partial linear model for skewed longitudinal data",
abstract = "Unlike majority of current statistical models and methods focusing on mean response for highly skewed longitudinal data, we present a novel model for such data accommodating a partially linear median regression function, a skewed error distribution and within subject association structures. We provide theoretical justifications for our methods including asymptotic properties of the posterior and associated semiparametric Bayesian estimators. We also provide simulation studies to investigate the finite sample properties of our methods. Several advantages of our method compared with existing methods are demonstrated via analysis of a cardiotoxicity study of children of HIV-infected mothers.",
keywords = "Dirichlet process, Median regression, Partial linear model, Semiparametric, Skewed error.",
author = "Yuanyuan Tang and Debajyoti Sinha and Debdeep Pati and Stuart Lipsitz and Lipshultz, {Steven E}",
year = "2014",
month = "9",
day = "17",
doi = "10.1093/biostatistics/kxv005",
language = "English (US)",
volume = "16",
pages = "441--453",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - Bayesian partial linear model for skewed longitudinal data

AU - Tang, Yuanyuan

AU - Sinha, Debajyoti

AU - Pati, Debdeep

AU - Lipsitz, Stuart

AU - Lipshultz, Steven E

PY - 2014/9/17

Y1 - 2014/9/17

N2 - Unlike majority of current statistical models and methods focusing on mean response for highly skewed longitudinal data, we present a novel model for such data accommodating a partially linear median regression function, a skewed error distribution and within subject association structures. We provide theoretical justifications for our methods including asymptotic properties of the posterior and associated semiparametric Bayesian estimators. We also provide simulation studies to investigate the finite sample properties of our methods. Several advantages of our method compared with existing methods are demonstrated via analysis of a cardiotoxicity study of children of HIV-infected mothers.

AB - Unlike majority of current statistical models and methods focusing on mean response for highly skewed longitudinal data, we present a novel model for such data accommodating a partially linear median regression function, a skewed error distribution and within subject association structures. We provide theoretical justifications for our methods including asymptotic properties of the posterior and associated semiparametric Bayesian estimators. We also provide simulation studies to investigate the finite sample properties of our methods. Several advantages of our method compared with existing methods are demonstrated via analysis of a cardiotoxicity study of children of HIV-infected mothers.

KW - Dirichlet process

KW - Median regression

KW - Partial linear model

KW - Semiparametric

KW - Skewed error.

UR - http://www.scopus.com/inward/record.url?scp=84936753513&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84936753513&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxv005

DO - 10.1093/biostatistics/kxv005

M3 - Article

C2 - 25792623

AN - SCOPUS:84936753513

VL - 16

SP - 441

EP - 453

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 3

ER -