Semiparametric model for the dichotomized functional outcome after stroke: The Northern Manhattan Study

Huaihou Chen, Myunghee Cho Paik, Mandip S. Dhamoon, Yeseon Park Moon, Joshua Willey, Ralph L Sacco, Mitchell S V Elkind

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The Northern Manhattan Study (NOMAS) is a prospective, population-based study. One of the goals of NOMAS is to characterize the functional status of stroke survivors over time after stroke. Based on generalized estimating equation models, previous parametric analysis showed that functional status declines over time and the trajectories of decline are different depending on insurance status. The two trends of functional status may not be linear, which motivates our semiparametric modeling. In this paper, we model the time trend nonparametrically, the associated covariates parametrically and an interaction term between the nonparametric time trend and a covariate. We consider both kernel weighted local polynomial-based and regression spline-based approaches for solving the semiparametric model, and propose a statistic to test for the interaction term. To evaluate the performance of the parametric model in the case of model misspecification, we study the bias and efficiency of the estimators from misspecified parametric models. We find that when the adjusted covariates are independent of the time, and the link function is identity, the estimators for those covariates are asymptotically unbiased, even if the time trend is misspecified. In general, however, under other conditions and nonidentity link, the misspecified parametric estimators are biased and less efficient even when they are unbiased. We compute the ARE and also conduct simulation studies and compare power for testing the adjusted covariate when the time trend is modeled parametrically versus nonparametrically. In the simulation studies, we observe significant gain in power of those semiparametric model-based estimators compared to the parametric model-based estimators in the cases when the time trend is nonlinear.

Original languageEnglish
Pages (from-to)2598-2608
Number of pages11
JournalComputational Statistics and Data Analysis
Volume56
Issue number8
DOIs
StatePublished - Aug 1 2012
Externally publishedYes

Fingerprint

Semiparametric Model
Stroke
Covariates
Estimator
Parametric Model
Simulation Study
Model-based
Misspecified Model
Regression Splines
Local Polynomial
Parametric Analysis
Model Misspecification
Link Function
Generalized Estimating Equations
Insurance
Term
Model Analysis
Interaction
Splines
Biased

Keywords

  • Generalized estimating equation
  • Kernel method
  • Regression splines
  • Semiparametric longitudinal data analysis

ASJC Scopus subject areas

  • Computational Mathematics
  • Computational Theory and Mathematics
  • Statistics and Probability
  • Applied Mathematics

Cite this

Semiparametric model for the dichotomized functional outcome after stroke : The Northern Manhattan Study. / Chen, Huaihou; Paik, Myunghee Cho; Dhamoon, Mandip S.; Moon, Yeseon Park; Willey, Joshua; Sacco, Ralph L; Elkind, Mitchell S V.

In: Computational Statistics and Data Analysis, Vol. 56, No. 8, 01.08.2012, p. 2598-2608.

Research output: Contribution to journalArticle

Chen, Huaihou ; Paik, Myunghee Cho ; Dhamoon, Mandip S. ; Moon, Yeseon Park ; Willey, Joshua ; Sacco, Ralph L ; Elkind, Mitchell S V. / Semiparametric model for the dichotomized functional outcome after stroke : The Northern Manhattan Study. In: Computational Statistics and Data Analysis. 2012 ; Vol. 56, No. 8. pp. 2598-2608.
@article{7edcdd934db6410f95d6fe5b356d1b83,
title = "Semiparametric model for the dichotomized functional outcome after stroke: The Northern Manhattan Study",
abstract = "The Northern Manhattan Study (NOMAS) is a prospective, population-based study. One of the goals of NOMAS is to characterize the functional status of stroke survivors over time after stroke. Based on generalized estimating equation models, previous parametric analysis showed that functional status declines over time and the trajectories of decline are different depending on insurance status. The two trends of functional status may not be linear, which motivates our semiparametric modeling. In this paper, we model the time trend nonparametrically, the associated covariates parametrically and an interaction term between the nonparametric time trend and a covariate. We consider both kernel weighted local polynomial-based and regression spline-based approaches for solving the semiparametric model, and propose a statistic to test for the interaction term. To evaluate the performance of the parametric model in the case of model misspecification, we study the bias and efficiency of the estimators from misspecified parametric models. We find that when the adjusted covariates are independent of the time, and the link function is identity, the estimators for those covariates are asymptotically unbiased, even if the time trend is misspecified. In general, however, under other conditions and nonidentity link, the misspecified parametric estimators are biased and less efficient even when they are unbiased. We compute the ARE and also conduct simulation studies and compare power for testing the adjusted covariate when the time trend is modeled parametrically versus nonparametrically. In the simulation studies, we observe significant gain in power of those semiparametric model-based estimators compared to the parametric model-based estimators in the cases when the time trend is nonlinear.",
keywords = "Generalized estimating equation, Kernel method, Regression splines, Semiparametric longitudinal data analysis",
author = "Huaihou Chen and Paik, {Myunghee Cho} and Dhamoon, {Mandip S.} and Moon, {Yeseon Park} and Joshua Willey and Sacco, {Ralph L} and Elkind, {Mitchell S V}",
year = "2012",
month = "8",
day = "1",
doi = "10.1016/j.csda.2012.02.001",
language = "English",
volume = "56",
pages = "2598--2608",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "8",

}

TY - JOUR

T1 - Semiparametric model for the dichotomized functional outcome after stroke

T2 - The Northern Manhattan Study

AU - Chen, Huaihou

AU - Paik, Myunghee Cho

AU - Dhamoon, Mandip S.

AU - Moon, Yeseon Park

AU - Willey, Joshua

AU - Sacco, Ralph L

AU - Elkind, Mitchell S V

PY - 2012/8/1

Y1 - 2012/8/1

N2 - The Northern Manhattan Study (NOMAS) is a prospective, population-based study. One of the goals of NOMAS is to characterize the functional status of stroke survivors over time after stroke. Based on generalized estimating equation models, previous parametric analysis showed that functional status declines over time and the trajectories of decline are different depending on insurance status. The two trends of functional status may not be linear, which motivates our semiparametric modeling. In this paper, we model the time trend nonparametrically, the associated covariates parametrically and an interaction term between the nonparametric time trend and a covariate. We consider both kernel weighted local polynomial-based and regression spline-based approaches for solving the semiparametric model, and propose a statistic to test for the interaction term. To evaluate the performance of the parametric model in the case of model misspecification, we study the bias and efficiency of the estimators from misspecified parametric models. We find that when the adjusted covariates are independent of the time, and the link function is identity, the estimators for those covariates are asymptotically unbiased, even if the time trend is misspecified. In general, however, under other conditions and nonidentity link, the misspecified parametric estimators are biased and less efficient even when they are unbiased. We compute the ARE and also conduct simulation studies and compare power for testing the adjusted covariate when the time trend is modeled parametrically versus nonparametrically. In the simulation studies, we observe significant gain in power of those semiparametric model-based estimators compared to the parametric model-based estimators in the cases when the time trend is nonlinear.

AB - The Northern Manhattan Study (NOMAS) is a prospective, population-based study. One of the goals of NOMAS is to characterize the functional status of stroke survivors over time after stroke. Based on generalized estimating equation models, previous parametric analysis showed that functional status declines over time and the trajectories of decline are different depending on insurance status. The two trends of functional status may not be linear, which motivates our semiparametric modeling. In this paper, we model the time trend nonparametrically, the associated covariates parametrically and an interaction term between the nonparametric time trend and a covariate. We consider both kernel weighted local polynomial-based and regression spline-based approaches for solving the semiparametric model, and propose a statistic to test for the interaction term. To evaluate the performance of the parametric model in the case of model misspecification, we study the bias and efficiency of the estimators from misspecified parametric models. We find that when the adjusted covariates are independent of the time, and the link function is identity, the estimators for those covariates are asymptotically unbiased, even if the time trend is misspecified. In general, however, under other conditions and nonidentity link, the misspecified parametric estimators are biased and less efficient even when they are unbiased. We compute the ARE and also conduct simulation studies and compare power for testing the adjusted covariate when the time trend is modeled parametrically versus nonparametrically. In the simulation studies, we observe significant gain in power of those semiparametric model-based estimators compared to the parametric model-based estimators in the cases when the time trend is nonlinear.

KW - Generalized estimating equation

KW - Kernel method

KW - Regression splines

KW - Semiparametric longitudinal data analysis

UR - http://www.scopus.com/inward/record.url?scp=84862826080&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862826080&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2012.02.001

DO - 10.1016/j.csda.2012.02.001

M3 - Article

AN - SCOPUS:84862826080

VL - 56

SP - 2598

EP - 2608

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 8

ER -