Skip Navigation Links
Link to the last RSS article here: Installing SPSS 14 - Ed.
Bayesian Packages for R version 2.2.1
By Dr Rich Herrington, ACS Research and Statistical Support Services Consultant
R version 2.2.1 was released on December 20, 2005. A description of changes for Microsoft Windows platform versions can be found here. The latest versions of R for the Windows platform can be found here. We are hosting a local copy of the R 2.2.1 installation file for the Windows platform here for download. Additionally, we have made free copies of a "executable" CD version of R 2.2.1 available over in the UNT bookstore (trade books). R 2.2.1 has been installed on this "live" CD with some minor pre-configurations so that R can run off of the CD. To begin the R session, browse to the \bin folder on the CD and click the Rgui.exe file. Additionally, you can set up a shortcut on your desktop to run R off of the CD. Alternatively, you can copy the contents of the CD into an R folder on your local hard-drive and create a shortcut to the C:\R\bin\Rgui.exe file. Once R has been started, you should see two windows once the initialization of R is complete. The first is the R Console window:
The second is the R Commander window, a simple drop down menu for doing a number of beginner to advanced statistics thru a GUI dialog system:
To set up your working directory so that you can read and write files to your local storage media (not the CD - since it is write only). Go the R console window and select "File - Change dir":
You can set your working directory to a writeable storage media:
Now you should be able to write out and read to this directory.
Bayesian Analysis in R
Bayesian approaches to inference have become increasingly popular in applied statistics since the arrival of cheap, fast, computers. The availability of today's computational power in a desktop PC allows more complicated, realistic Bayesian models to be estimated thru simulation methodologies. Currently, the availability of the BUGS software (e.g. WinBugs & OpenBugs) and numerous R packages dedicated to Bayesian analysis, give the researcher an arsenal of methods to attack problems from a Bayesian framework. An introduction to Bayesian thinking and data analysis is beyond the scope of this current article, but we hope to cover an introduction to Bayesian analysis in R and WinBugs in future column installments. Here, we just present a sampling of some of the packages that are available in R for doing Bayesian analysis:
R2WinBUGS Running WinBUGS from R
A Few Good Books on Bayesian Analysis
I'll end this column by listing a few of my favorite books on Bayesian Analysis:
Bayesian Statistics and Marketing