This preview shows pages 1–11. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: S t a t i s t i c s Ucla Jan de Leeuw Computational Statistics with R Part I: R Stat 135 What is R ? R Resources R Session R Objects 2 Table of Contents This course will discuss  programming in R, computational statistics in R. The general idea is to work through a number of projects. On the computational statistics side we will discuss floating point, matrix computations, optimization, and Monte Carlo. On the R side we will discuss R interfaces, R resources, R packages, R symbols, R sessions, R functional programming, R I/ O, using C in R and R in C. 3 Stat 135 The website for this course is http://www.cuddyvalley.org/teaching/135 This has the latest version of the handouts in pdf and keynote, as well as supplementary material (code and documentation). The mailing list is stat135@cuddyvalley.org and you can subscribe to this list at http://www.cuddyvalley.org/mailman/listinfo/ stat135 4 R is a software environment for statistical computing and graphics. R is free, open source, and available in binary form for Macintosh, Windows, and most Unix systems. R development is managed by Rcore, a team of about 20 volunteers. The homepage of the R project is http://www.r project.org . 5 What is R ? Initially, there was S, created in the early eighties at Bell Labs, eventually commercialized in the nineties as SPlus. One can think of S as the language used in both R and SPlus. Or as R and SPlus as two different implementations of S. There are some differences between R and S Plus. Since the balance is rapidly shifting from S Plus to R, we do not discuss Splus. 6 R is based (internally) on Scheme, a Lisp dialect. R is constructed to be compatible with software written in S. In fact, R can be thought of as the open source GNU version of S. The real power of R comes from the fact that it is developed actively, that various convenient GUIs are available, that the whole statistics community can and does contribute to the growth of R, that publishing companies such as Springer have adopted R, and that R is now the lingua franca of Statistics. 7 You can run R from a GUI (Windows, Mac, Gnome, Tcl/Tk, Rcmdr) or from a terminal window (telnet, xterm, emacs). R can generate graphics in pdf, jpeg, png, ps (files) and or in X11, Aqua, Windows (windows). R can be downloaded (as binary or as source code) from http://cran.stat.ucla.edu . 8 R resources 9 Windows, OS X, Unix Emacs, Aqua, Windows, Xterm, Terminal.app R R packages CRAN has between 500 and 1000 addon packages. There are hundreds more packages on other sites, such as the Bioconductor site http:// www.bioconductor.org . Some packages are required , some are recommended , some are optional . All of them taken together are a bit of a zoo, with a great deal of overlap and a great deal of variation in size, importance, and quality....
View
Full
Document
 Fall '10
 JandeLeeuw
 Statistics

Click to edit the document details