Supplement 1. R software packages for boosted trees.

File List gbmplus_1.5-17.zip R package suitable for installation on Windows OS. gbmplus_1.5-17.tar.gz R package sources suitable for installation on all supported OS(Linux/Unix/Windows). softcorals.csv Data for soft coral analysis in paper. Variable names are in row 1 and details of all variables are supplied in the paper. barramundi.csv Data for barramundi analysis in Appendix One. Variable names are in row 1 and details of all variables are supplied in Appendix One. Description Package: gbmplus Version: 1.5-17 Date: 2006-07-01 Title: Generalized Boosted Regression Models Author: Greg Ridgeway Modifications of package gbm by Glenn De'ath Maintainer: Glenn De'ath Depends: R (>= 2.1.0), lattice Description: This package implements extensions to Freund and Schapire's AdaBoost algorithm and J. Friedman's gradient boosting machine. Includes regression methods for least squares, absolute loss, logistic, Poisson, Cox proportional hazards partial likelihood, and AdaBoost exponential loss. Includes aggregated boosting (GD). License: GPL (version 2 or newer) URL: http://www.i-pensieri.com/gregr/gbm.shtml Packaged: Tue Jul 4 11:34:54 2006; Glenn The most flexible, accessible environment with access to tree and boosting software is the R software. R is a freely available GNU version of the S language, includes extensive statistical computing and graphics procedures, and runs on a wide variety of UNIX platforms, Windows and MacOS. R packages for fitting classification and regression trees include rpart, mvpart and tree. For boosting, the packages gbm and boost are available and gbmplus (a modified version of gbm) that implements aggregated boosting and is available from here or the author at [e-mail]. The R packages randomForest and ipred are available for random forests and bagging. Commercial software for trees, boosting and random forests is available from Salford Systems. The package gbmplus is available from these Ecological Archives in zip format for Windows users and tar format for Linux users.