A Semiparametric Bayesian Approach to Multivariate Longitudinal Data.

Pulak Ghosh, Timothy Hanson
Author Information
  1. Pulak Ghosh: Department of Quantitative Methods & Information Systems, Indian Institute of Management, Bangalore, India.

Abstract

We extend the standard multivariate mixed model by incorporating a smooth time effect and relaxing distributional assumptions. We propose a semiparametric Bayesian approach to multivariate longitudinal data using a mixture of Polya trees prior distribution. Usually, the distribution of random effects in a longitudinal data model is assumed to be Gaussian. However, the normality assumption may be suspect, particularly if the estimated longitudinal trajectory parameters exhibit multimodality and skewness. In this paper we propose a mixture of Polya trees prior density to address the limitations of the parametric random effects distribution. We illustrate the methodology by analyzing data from a recent HIV-AIDS study.

References

  1. JAMA. 2002 Jul 10;288(2):169-80 [PMID: 12095381]
  2. Biometrics. 2001 Sep;57(3):795-802 [PMID: 11550930]
  3. Biometrics. 2006 Jun;62(2):424-31 [PMID: 16918906]
  4. Biometrics. 1984 Dec;40(4):1079-87 [PMID: 6534410]
  5. Biometrics. 1982 Dec;38(4):963-74 [PMID: 7168798]
  6. Biometrics. 2003 Jun;59(2):221-8 [PMID: 12926706]
  7. Biometrics. 1999 Jun;55(2):477-83 [PMID: 11318203]
  8. Biometrics. 2002 Dec;58(4):742-53 [PMID: 12495128]
  9. Biometrics. 2004 Dec;60(4):945-53 [PMID: 15606415]
  10. Biometrics. 1969 Jun;25(2):357-81 [PMID: 5794105]
  11. Biometrics. 1994 Sep;50(3):689-99 [PMID: 7981395]
  12. Biometrics. 2005 Mar;61(1):64-73 [PMID: 15737079]
  13. Biometrics. 1998 Sep;54(3):921-38 [PMID: 9750242]

Grants

  1. R01 CA095955/NCI NIH HHS
  2. R01 CA095955-05A2/NCI NIH HHS

Word Cloud

Created with Highcharts 10.0.0longitudinaldatadistributionmultivariatemodelproposeBayesianmixturePolyatreespriorrandomeffectsextendstandardmixedincorporatingsmoothtimeeffectrelaxingdistributionalassumptionssemiparametricapproachusingUsuallyassumedGaussianHowevernormalityassumptionmaysuspectparticularlyestimatedtrajectoryparametersexhibitmultimodalityskewnesspaperdensityaddresslimitationsparametricillustratemethodologyanalyzingrecentHIV-AIDSstudySemiparametricApproachMultivariateLongitudinalData

Similar Articles

Cited By (6)