The oem package for penalized regression is on CRAN

The oem package has been on CRAN for some time now, but with the latest update I expect few structural changes to the user interface. oem is a package for the estimation of various penalized regression models using the oem algorithm of Xiong et al. (2016). The focus of oem is to provide high performance computation for big tall data. Many applications not only have a large number of variables, but a vast number of observations; oem is designed to perform well in these settings.

Fast and Big Linear Model Fitting with bigmemory and RcppEigen

In a previous post, I went over the basics of linking up bigmemory and the eigen C++ library via RcppEigen. In this post I’ll take this a bit further by creating a version of the fastLm() function of RcppEigen that can accept bigmemory objects. By doing so, we will create a fast way to fit linear models using data which is too big to fit in RAM. With RcppEigen, fitting linear models using out-of-memory computation doesn’t have to be slow. The code for this is all on github in the bigFastlm package.