Statistics, Department of

 

The R Journal

Date of this Version

6-2019

Document Type

Article

Citation

The R Journal (June 2019) 11(1); Editor: Norm Matloff

Comments

Copyright 2019, The R Foundation. Open access material. License: CC BY 4.0 International

Abstract

Though longitudinal data often contain missing responses and error-prone covariates, relatively little work has been available to simultaneously correct for the effects of response missingness and covariate measurement error on analysis of longitudinal data. Yi (2008) proposed a simulation based marginal method to adjust for the bias induced by measurement error in covariates as well as by missingness in response. The proposed method focuses on modeling the marginal mean and variance structures, and the missing at random mechanism is assumed. Furthermore, the distribution of covariates are left unspecified. These features make the proposed method applicable to a broad settings. In this paper, we develop an R package, called swgee, which implements the method proposed by Yi (2008). Moreover, our package includes additional implementation steps which extend the setting considered by Yi (2008). To describe the use of the package and its main features, we report simulation studies and analyses of a data set arising from the Framingham Heart Study

Download

Share

COinS