R: An Overview and Some Current Directions

Luke Tierney

R is an open source language for statistical computing and graphics basedon the ACM software award-winning S language. R is widely used for dataanalysis and has become a major vehicle for making available new statisticalmethodology. This paper presents an overview of the design philosophy andthe development model for R, reviews the basic capabilities of the system,and outlines some current projects that will inuence future developmentsof R.AMS 2000 subject classications.Primary 68N15; Secondary 62-07.Keywords.Computer language, graphics, isosurfaces, parallel computation, R, statisticalcomputing.1. IntroductionR is a language for statistical computing and graphics. R, along with S-plus,is a member of the S language family originally developed by John Chambers andcolleagues at Bell Laboratories (Becker and Chambers, 1984; Beckeret al., 1988;Chambers, 1998). In 1998 the Association for Computing Machinery presentedits prestigious Software System Award to John Chambers for his development ofthe S system. In recent years the S language family, and R in particular, havebecome the de facto standard for computing in statistical research. Many booksdescribing R and S are now available; two examples are Dalgaard (2004) andVenables and Ripley (2002).R can viewed as a dierent implementation or as a dialect of S the language.While there are some important dierences between R and S-plus, some of whichare discussed in Section 6.1, much code written for S-plus runs unaltered underR. In recent years R has become a major vehicle for making available new sta-tistical methodology. Articles proposing new methods that also make availableReceived October 2006; accepted November 2006.†This research supported in part by National Sciance Foundation Grant DMS-0604593. KCI Citation Count: 1

R: An Overview and Some Current Directions

View Online

Abstract

Details

Metrics