Penalized Generalized Estimating Equations for High‐Dimensional Longitudinal Data Analysis

We consider the penalized generalized estimating equations (GEEs) for analyzing longitudinal data with high‐dimensional covariates, which often arise in microarray experiments and large‐scale health studies. Existing high‐dimensional regression procedures often assume independent data and rely on th...

Full description

Saved in:
Bibliographic Details
Published in:Biometrics 2012-06, Vol.68 (2), p.353-360
Main Authors: Wang, Lan, Zhou, Jianhui, Qu, Annie
Format: Article
Language:English
Subjects:
Algorithms
BIOMETRIC METHODOLOGY
Biometrics
biometry
Biometry - methods
cell cycle
Cell Cycle - genetics
Coefficients
Computer Simulation
Correlated data
Correlation analysis
Data analysis
data collection
Data Interpretation, Statistical
Diverging number of parameters
equations
Estimate reliability
Estimating techniques
Estimators
GEE
Gene expression
Genes
High-dimensional covariates
Humans
Likelihood Functions
Longitudinal data
Longitudinal Studies
Marginal regression
Maximum likelihood method
microarray technology
Models, Statistical
Monte Carlo Method
Oracles
probability
Saccharomyces cerevisiae - cytology
Saccharomyces cerevisiae - genetics
Standard deviation
variable selection
Yeasts
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We consider the penalized generalized estimating equations (GEEs) for analyzing longitudinal data with high‐dimensional covariates, which often arise in microarray experiments and large‐scale health studies. Existing high‐dimensional regression procedures often assume independent data and rely on the likelihood function. Construction of a feasible joint likelihood function for high‐dimensional longitudinal data is challenging, particularly for correlated discrete outcome data. The penalized GEE procedure only requires specifying the first two marginal moments and a working correlation structure. We establish the asymptotic theory in a high‐dimensional framework where the number of covariates pn increases as the number of clusters n increases, and pn can reach the same order as n. One important feature of the new procedure is that the consistency of model selection holds even if the working correlation structure is misspecified. We evaluate the performance of the proposed method using Monte Carlo simulations and demonstrate its application using a yeast cell‐cycle gene expression data set.
ISSN:0006-341X
1541-0420
DOI:10.1111/j.1541-0420.2011.01678.x