R Statistical Computing Environment
We will primarily be using the R
statistical computing package in this course. R is a freely
available, open-source version of S/S-plus. While most S/S-plus code
will run in R, there are several differences. Therefore, it is
recommended that all computing for the course be done in R.
Previous experience with statistical computing is a prerequisite for
this course. However, knowledge of R or S/S-plus is not expected.
Demonstrations of analyses in R will be given during lecture, and
example code will be provided for lecture examples as well as with
some homework assignments. In addition to this example code, it is
highly recommended that the R manual
and online help system be used as a supplement.
Students not familiar with R may want to work through an R tutorial prepared by Hongfei Li. Data sets used in
the tutorial can be downloaded here: toxicity.dat, lakes.dat
R is available in the Department of Statistics Student Computer
Laboratory, which is open only to Statistics students. R can also be
to personal computers for no cost. Students who do not have access to
the computer lab and have difficulty accessing R should speak to the
Students using R on their personal computers may
want to download Tinn-R,
an R code editor -- see the installation and configuration instructions
(prepared by Dale Rhoda).
NOTE: You do not need to use Tinn-R to
edit R code; you can use Notepad, as well as other text editors
available in Windows.
In addition to R, we will be using WinBUGS, a
software package for fitting complex statistical models using Markov
chain Monte Carlo (MCMC) methods. In particular, we will be using a
freely available, open-source version of WinBUGS, called OpenBUGS. OpenBUGS
can be called directly from R using the BRugs
R package. OpenBUGS software is also available in the Department of
Statistics Student Computer Laboratory and can be downloaded
to students' personal computers. An overview of WinBUGS and many
examples can be found in the WinBUGS