Variable selection for longitudinal survey data

Authors

  • Laura Dumitrescu School of Mathematics and Statistics, Victoria University of Wellington, Wellington 6140, New Zealand
  • Wei Qian Statistics Canada, 100 Tunney’s Pasture Driveway, Ottawa, K1A 0T6, Canada
  • JNK Rao School of Mathematics and Statistics, Carleton University, Ottawa, K1S 5B6, Canada

Keywords:

Complex sampling design; longitudinal data; model selection; oracle property; quadratic inference functions; SCAD; super-population model.

Abstract

In this article we propose a new variable selection method for analyzing data collected from longitudinal sample surveys. The procedure is based on the survey-weighted quadratic inference function, which was recently introduced as an alternative to the survey-weighted generalized estimating function. Under the joint model-design framework, we introduce the penalized survey-weighted quadratic inference estimator and obtain sufficient conditions for the existence, weak consistency, sparsity and asymptotic normality. To illustrate the finite sample performance of the model selection procedure, we include a limited simulation study.

Journal of Statistical Research 2021, Vol. 55, No. 1, pp. 21-41

Abstract
56
PDF
46

Downloads

Published

2021-12-09

How to Cite

Dumitrescu, L. ., Qian, W. ., & Rao, J. . (2021). Variable selection for longitudinal survey data. Journal of Statistical Research, 55(1), 21–41. Retrieved from https://banglajol.info/index.php/JStR/article/view/56565

Issue

Section

Articles