Useful Links

This page contains various links to the web resources mentioned on the Numerical Analysis of Biological Data course, run by the ECRC in London.

Canoco Version 4.5 for Windows
Canoco is distributed by Microcomputer Power. Their web pages contain all the information you will need to purchase the Canoco software. Canoco 4.5 now includes a Canodraw for Windows, offering a simple interface to basic and advanced graphics.

Canoco also has an official home page @ Plant Research international.

Mike Palmer's Ordination web pagesThese pages are a very useful and thorough resource for ordination methods. Mike's pages also have a series of links to software on the web that perform numerical analyses of biological data, as well as links to other web sites that deal with similar methods.

The ORDNEWS Listserver You can post your questions about ordination methods and other numerical methods to this friendly listserver. This really is a useful resource, even if you just read the postings made by other people and the replies they receive.

CCA Bibliography This is an online list of references that use CCA. Compiled by John Birks and Sylvia Peglar, and maintained by Phillippe Casgrain in Montreal.

Pierre Legendre's Web Pages
Here you can find information about Pierre's work, publications and freely available software.

Steve Juggins' Web Pages
These pages are home to various programs written by Steve. You can download many of them direct from the web site, and manuals are available from here as well.

Jari Oksanen
Jari's web pages describe his interests in numerical ecology. You can download an up-to-date version of the HOF program for fitting hierarchical models as species response curves, as well as other software he has written. Jari has also written a package of functions for the R statistical language called vegan, and a pdf appendix to Oksanen and Minchin (2002) shows how to fit HOF models in R and compares them to GAM and beta response models.

The R-Package
This is a freely available package of programs for performing a huge range of multivariate analyses, and is NOT to be confused with R (see below). The current (non-beta) release is version 3 and runs on a Macintosh computer. Work is progressing on the vastly updated Version 4. This is currently in alpha or beta mode (i.e. not released) and only the Mac version is available. A windows version is planned to follow once the Mac version is up and running. Certainly something to keep an eye on!

Philippe Casgrain has announced that the Windows version of the R-Package (V4.0) now has its own web page where you can follow developments. He has also set up an announce list that you can subscribe to. This list will be used to announce the release of the Windows version and updates thereafter.

The R statistical language
Throughout the course John made reference to a program called S-Plus that is capable of performing the more advanced methods (e.g. GAM's, C&RTS etc). S-Plus is a commercial package based on the S computer language and is available from Insightful. As such you have to pay for it! There is an open source project called the R-Project (not to be confused with the R-Package). R is an environment and language for performing statistics. It is free and can do most of the things S-Plus can, and in many cases it does them better! It doesn't have a "proper" GUI, so S-Plus is perhaps easier to get to grips with quickly.

R is very powerful! It is free! It is well supported by a community of statisticians and non-statisticians from a huge range of fields! But will take a bit of getting used to.

There are now many packages of functions that have been created which specifically deal with ecological methods; including vegan, LabDSV, ade4, and Brian McArdle's MultivEcol. Or you can write your own functions based on what you have learned this fortnight!

Statistica
Statistica is made by StatSoft and details of the software can be found on their web site.

StatSoft also have an online "Electronic Statistics Textbook". It is very comperhensive and a useful resource for statistical techniques even if you don't use Statistica. As you would expect, it is biased towards things in Statistica though.

References:
Oksanen, J., Minchin, P. R., 2002. Continuum theory revisited: what shape are
species responses along ecological gradients? Ecological Modelling Volume 157, Issues 2-3, Pages 131-139