R package “ELCIC”

We developed a consistent and robust information criterion to conduct model selection for semiparametric models. It is free of distribution specification and powerful to locate the true model given large sample size. This package provides several usage of ELCIC with applications in generalized linear model (GLM), generalized estimating equation (GEE) for longitudinal data, and weighted GEE (WGEE) for missing longitudinal data under the mechanism of missing at random and drop-out. View more details on https://cran.r-project.org/web/packages/ELCIC/index.html.

You can also refer to the paper:“Chen C, Shen B and Wang M. ELCIC: An R package for model selection using the empirical-likelihood based information criterion. Communications for Statistical Applications and Methods, 2023. Accepted and to appear.”.