PGEE: An R Package for Analysis of Longitudinal Data with High-Dimensional Covariates

Inan G., Wang L.

R JOURNAL, vol.9, no.1, pp.393-402, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 9 Issue: 1
  • Publication Date: 2017
  • Journal Name: R JOURNAL
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.393-402
  • Istanbul Technical University Affiliated: No

Abstract

We introduce an R package PGEE that implements the penalized generalized estimating equations (GEE) procedure proposed byWang et al. (2012) to analyze longitudinal data with a large number of covariates. The PGEE package includes three main functions: CVfit, PGEE, and MGEE. The CVfit function computes the cross-validated tuning parameter for penalized generalized estimating equations. The function PGEE performs simultaneous estimation and variable selection for longitudinal data with high-dimensional covariates; whereas the function MGEE fits unpenalized GEE to the data for comparison. The R package PGEE is illustrated using a yeast cell-cycle gene expression data set.