We are releasing ncpen R package

ncpen: A Model Selection Algorithm for High-dimensional Datasets By Dongshin (Donn) Kim, Sunghoon Kwon and Sangin Lee

I recently published a statistical R software package, ncpen, with my two statistician co-authors. For a data set with many variables (high-dimensional data), like over 1,000 or 10,000 and more, this algorithm selects only relevant variables producing a parsimonious regression model. This package works well with big-data. Some of you maybe heard about LASSO (least absolute shrinkage and selection operator). ncpen offers similar functions to LASSO but has a lot better characteristics for economics and finance research. If you are an R user, go and try install.packages("ncpen").

More details are available at our ncpen project website. The package is fully functioning. You can check out ncpen manual and ncpen usage example. This is also available at CRAN (official R depository), ncpen package. Since this is an initial release, we have more things to do for the documentation. At the moment, we are working on a paper about this package. Results are great! Your comments are always welcome.