A cocktail algorithm for solving the elastic net penalized cox's regression in high dimensions

Abstract

We introduce a cocktail algorithm, a good mixture of coordinate decent, the majorization-minimization principle and the strong rule, for computing the solution paths of the elastic net penalized Cox's proportional hazards model. The cocktail algorithm enjoys a proven convergence property. We have implemented the cocktail algorithm in an R package fastcox. Numerical examples show that cocktail is comparable to coxnet [11] in speed and often delivers better quality solutions.