Rlanguage functions

Anova and ttests
 anova.1way.R (P. Legendre): Oneway anova with permutation test.
 anova.2way.R (P. Legendre): Twoway crossedfactor anova with permutation tests (balanced design): models I, II, and III.
 anova.2way.unbalanced.R (P. Legendre): 2way multivariate anova (by RDA) for balanced or unbalanced designs for 2 fixed crossed factors, with permutation tests.
 anova.3way.R (P. Legendre): Threeway crossedfactor anova (fixed factors) with permutation tests (balanced design).
 nest.anova.perm.R (D. Borcard and P. Legendre): Nested anova with permutation tests (main factor and one nested factor, balanced design).
 t.perm.R (P. Legendre): ttest for independent samples with permutation test.
 t.paired.perm.R (P. Legendre & G. Blanchet): ttest for related samples with permutation test.

beta.div (four functions and documentation)
 beta.div (P. Legendre): This function computes estimates of the total beta diversity as the total variance in a community data matrix Y, for 17 dissimilarity coefficients or the raw data table. It also computes the species contributions (SCBD) and the local contributions (LCBD) to beta diversity.
 beta.div.comp (P. Legendre): Podani and Baselga decompositions of a dissimilarity matrix into Replacement and Richness/Abundance difference components. The function can handle species presenceabsence or abundance data, or biomasses per species. See the Legendre (2014) paper. Appendix 5 of that paper contains a function to compute LCBD indices from a Dissimilarity, Replacement, or Richness/Abundance difference matrix.
 LCBD.comp (P. Legendre): This function computes local contributions to beta diversity (LCBD indices) from a symmetric dissimilarity matrix (D) or from a beta component matrix (Repl, RichDiff or AbDiff, or Nes).
 TBI (P. Legendre): This function computes the differences between multivariate observations (frequency or presenceabsence data) forming pairs observed at times T1 and T2. Temporal Betadiversity Indices (TBI) are computed and tested. TBI are dissimilarity indices that measure beta differentiation through time. They are computed separately between T1 and T2 for each site.

bgdispersal (P. Legendre): Coefficients of biogeographical dispersal direction. Incorporated into the vegan library.

broken.stick.R (P. Legendre): Computes the expected values of the brokenstick distribution.

CADM: (P. Legendre): Concordance among several distance matrices. This function is now available in the ape library for phylogenetic analysis.

Canonical correspondence analysis (CCA)
 CCA (P. Legendre): Demonstration function for canonical correspondence analysis (ter Braak 1986, 1987), following the algorithm described in Subsection 11.2.1 of Legendre and Legendre (1998).

Canonical redundancy analysis (RDA)

rdaTest (P. Legendre and S. Durand): R library to compute simple and partial canonical redundancy analysis (RDA) with permutation test and draw triplots of the results. Older versions of these functions are also found in the library sonarX, which is distributed on the ESA Web page http://esapubs.org/archive/, inside “Ecological Archives A016047S1”. R library modified April 2014.
 AdjustedRsquare.R (P. Legendre): This function computes adjusted Rsquares for two types of regression.
 dbRDA.D.R (P. Legendre): This function computes the dbRDA Ftest of significance for distance matrices that have, or not, the Euclidean property, following the algorithm of McArdle & Anderson (2001). The Fstatistic is obtained without computation of the eigenvalues and eigenvectors, hence no correction has to be made to the D matrix to eliminate the negative eigenvalues.


cascadeKM (S. Durand, M.H. Ouellette and P. Legendre): Wrapper for the kmeans function for Kmeans partitioning. It creates several partitions forming a cascade from a small to a large number of groups. Incorporated into the vegan library.
Also distributed as part of the “Ecological Archives A016047S1” on the ESA Web page http://esapubs.org/archive/appl/A016/047/. 
CCorA (P. Legendre): Canonical and partial canonical correlation analysis, Y <> X1/X2. A version of that function for the calculation of simple canonical correlation analysis is found in the vegan library.

const.clust (P. Legendre): R package to compute spaceconstrained or timeconstrained agglomerative clustering from a dissimilarity matrix computed from multivariate data. The different solutions are tested by crossvalidation. The clustering results are presented in the form of maps. R library modified March 30th, 2014.
 Versions for R 3.0
 const.clust , crossplatform source
 const.clust library for Windows (R 3.01)
 const.clust library for Mac (R 3.01)
 Versions for R 2.12
 const.clust library for Windows (tested on R 2.12)
 const.clust library for Mac (R 2.12)
 Versions for R 3.0

corPerm.R (P. Legendre): Three functions to test the Pearson correlation coefficient by permutation.

dagnelie.test.R (D. Borcard, P. Legendre): Pierre Dagnelie’s test of multinormality.

Forward selection of explanatory variables in multiple regression and redundancy analysis (RDA)
 packfor (S. Dray, P. Legendre and G. Blanchet): Compiled package containing the functions “forward.sel” and “forward.sel.par”.
 R 3.0x Mac OSX download them here
 AEM library for Mac OSX (R 3.0x)
 packfor library for Mac OSX (R 3.0x)
 PCNM library for Mac OSX (R 3.0x)
 spacemakeR library for Mac OSX (R 3.0x)
 R 3.0x Mac OSX download them here
 packfor (S. Dray, P. Legendre and G. Blanchet): Compiled package containing the functions “forward.sel” and “forward.sel.par”.

fourthcorner (S. Dray): Fourthcorner analysis.

Functional diversity: FD library (Étienne Laliberté)

hclust.PL (P. Legendre): New version of the function “hclust” of {stats}, where the Fortran subroutine hclust.f has been recoded in R language for easy modification. This version is used in the followig paper: Murtagh, F. and P. Legendre “Ward’s hierarchical agglomerative clustering method: which algorithms implement Ward’s criterion?” (submitted paper).

Indicator / diagnostic species analyses: ‘indicspecies’ library: R package section.

Kendall.W: Species associations through Kendall’s coefficient of concordance. Functions kendall.global() et kendall.post() forming this library are now integrated into the vegan library for community analysis.

Linear regression
 multRegress.R (P. Legendre): This function computes a multiple regression and tests the coefficient of determination (Rsquare) by permutation.
 lmodel2 (P. Legendre): Package to compute model II simple linear regression using ordinary least squares (OLS), major axis (MA), standard major axis (SMA), and ranged major axis (RMA). Available on the CRAN page.
 lmorigin (P. Legendre): Package to compute multiple linear regressions with permutation tests. The program includes regression through the origin, to be used in particular for the analysis of independent contrast data. This function is now available in the ape library for phylogenetic analysis.

manova.2way.unbalanced.R (P. Legendre): manova for univariate or multivariate response data, two crossed factors, balanced or unbalanced design.

manovRDa.R (E. Laliberté): Twoway Manovalike RDA for fixed or random factors.

mantel.correlog (P. Legendre): Package to compute a multivariate Mantel correlogram. This function is now available in the vegan library for community analysis.

parafit (P. Legendre): Package to test hostparasite coevolution. This function is now available in the ape library for phylogenetic analysis.

PCNM, MEM and AEM spatial eigenfunctions
 PCNM function: The folder contains R functions PCNM and pcoa.all, documentation files (in pdf) for these two functions, as well as a page of instructions describing how to use these functions in the R console.

Periodograms
 Periodograph (P. Legendre): Contingency periodogram

WhittakerRobinson periodogram (P. Legendre): Periodic analysis of univariate quantitative data series.
New, faster version, based on C code, compiled for Mac OSX and Windows platforms in April 2014.

seriation.R (P. Legendre and S. Durand ): Seriation method of Beum & Brundage (1950) for nonsymmetric or symmetric resemblance matrices

sonarX (S. Durand et P. Legendre): Library for the extraction and analysis of singlebeam sonar signals for benthic habitat detection. Distributed as part of the “Ecological Archives A016047S1” on the ESA Web page http://esapubs.org/archive/appl/A016/047/.

Sidak.R (P. Legendre): Bonferroni and Sidak corrections for multiple testing.

Simple ordinations

sonarX (S. Durand and P. Legendre): Extraction and analysis of singlebeam sonar signals for benthic habitat detection and mapping. Package distributed as part of the “Ecological Archives A016047S1” on the ESA Web page http://esapubs.org/archive/appl/A016/047/.

STI: test of spacetime interaction in sampling design without replication: library distributed on the Software page of Miquel De Cáceres at http://vegmod.ctfc.cat

varpart (P. Legendre): Partition the variation of response table Y with respect to 2, 3, or 4 tables of explanatory variables. Incorporated into the vegan library.
 Ten years of vegan (Jari Oksanen): A short history of the development of the vegan package before and since its release on CRAN on September 6, 2001.