The R Project

In my copious free time (yes, that is a sarcastic use of "copious"), I copyedit for Limnology and Oceanography: Methods via ScholarOne, a subsidiary of Thomson Scientific. Recently I was working on the article, Karel Van den Meersche, Karline Soetaert, Jack J. Middelburg
A Bayesian Compositional Estimator for microbial taxonomy based on biomarkers (preview of abstract available at http://www.aslo.org/lomethods/free/2008/index.html). Of course, I cannot tell you more about this article, but because Eckerd subscribes, you will be able to access it once it is published. However, reading this article, I learned about the R Project for Statistical Computing. According to its home page, "R is a free software environment for statistical computing and graphics." Given that it is free, I wanted the NAS community to be aware of this software. The article mentioned above, when published, will demonstrate one use. Eckerd students and faculty may want to investigate other uses for their own research.

Statistics Library

Started in 1989 as an email service, StatLib is evolving into a powerful Internet tool. Hosted by Carnegie Mellon University and maintained by Pantelis Vlachos, a research scientist in Carnegie Mellon's Department of Statistics, StatLib contains freely available software and data sets for statisticians. The majority of both the software and data sets are affiliated with a published papers or books. For example, one dataset is described as

A collection of the data sets used in the book Analyzing Categorical Data, by Jeffrey S. Simonoff, Springer-Verlag, New York, 2003.

To fully understand the context and/or development of the resources, articles can be obtained from the library or Interlibrary Loan. Software is divided mainly by method, such as MATLAB extensions, multivariate archive, and more.

In addition, the site provides access to the Data and Story Library (DASL). Searchable by topics and statistical method used, DASL offers stories that are linked to data files.

Stories are abstracts that discuss the statistical concepts of a particular datafile. (from DASL)