SIS: An R Package for Sure Independence Screening in Ultrahigh-Dimensional Statistical Models

• Main Authors: Diego Franco Saldana, Yang Feng
• Format: Article
• Language: English
• Published: Foundation for Open Access Statistics 2018-02-01
• Series: Journal of Statistical Software
• Subjects: Cox model; generalized linear models; penalized likelihood estimation; sparsity; sure independence screening; variable selection
• Online Access: https://www.jstatsoft.org/index.php/jss/article/view/3378

We revisit sure independence screening procedures for variable selection in generalized linear models and the Cox proportional hazards model. Through the publicly available R package SIS, we provide a unified environment to carry out variable selection using iterative sure independence screening (ISIS) and all of its variants. For the regularization steps in the ISIS recruiting process, available penalties include the LASSO, SCAD, and MCP while the implemented variants for the screening steps are sample splitting, data-driven thresholding, and combinations thereof. Performance of these feature selection techniques is investigated by means of real and simulated data sets, where we find considerable improvements in terms of model selection and computational time between our algorithms and traditional penalized pseudo-likelihood methods applied directly to the full set of covariates.