In grpreg: Regularization Paths for Regression Models with Grouped Covariates

Iteratively reweighted algorithms introduce scaling issues for nonconvex penalties. In grpreg, this is resolved through a technique called adaptive rescaling.

This issue is documented in more detail on the ncvreg webpage and was originally proposed in the following manuscript:

• Breheny P and Huang J (2011). Coordinate descent algorithms for nonconvex penalized regression, with applications to biological feature selection. Annals of Applied Statistics, 5: 232--253

grpreg documentation built on Sept. 11, 2024, 5:13 p.m.