Description
High-Dimensional Screening for Semiparametric Longitudinal Regression.
Description
Implements variable screening techniques for ultra-high dimensional regression settings. Techniques for independent (iid) data, varying-coefficient models, and longitudinal data are implemented. The package currently contains three screen functions: screenIID(), screenLD() and screenVCM(), and six methods for simulating dataset: simulateDCSIS(), simulateLD, simulateMVSIS(), simulateMVSISNY(), simulateSIRS() and simulateVCM(). The package is based on the work of Li-Ping ZHU, Lexin LI, Runze LI, and Li-Xing ZHU (2011) <DOI:10.1198/jasa.2011.tm10563>, Runze LI, Wei ZHONG, & Liping ZHU (2012) <DOI:10.1080/01621459.2012.695654>, Jingyuan LIU, Runze LI, & Rongling WU (2014) <DOI:10.1080/01621459.2013.850086> Hengjian CUI, Runze LI, & Wei ZHONG (2015) <DOI:10.1080/01621459.2014.920256>, and Wanghuan CHU, Runze LI and Matthew REIMHERR (2016) <DOI:10.1214/16-AOAS912>.