RdaCca

RdaCca is a handy, easy to use application specially designed to perform two forms of canonical analysis, for example redundancy analysis and canonical correspondence analysis, using multiple regression followed by direct eigenanalysis.

The method of calculation is described in Chapter 11 of Legendre & Legendre. This program is a pedagogical tool. Its objective is to demonstrate that the calculations can indeed be carried out in this way, and to provide users with a simple program allowing one to obtain all the eigenvectors and ordination axes from PCA or CA ordination, or from canonical analysis (RDA or CCA).

It is not meant to replace advanced programs for canonical ordination, such as CANOCO. In particular, this program does not do forward selection of explanatory variables, nor does it carry out permutation tests of significance. Eigenanalysis is carried out using either a Householder procedure or singular value decomposition (SVD). The subroutines used are from Numerical Recipes.