[Vegan-commits] r2150 - in pkg/vegan: R inst man

[noreply at r-forge.r-project.org]

Author: jarioksa
Date: 2012-05-02 07:55:40 +0200 (Wed, 02 May 2012)
New Revision: 2150

Modified:
   pkg/vegan/R/monoMDS.R
   pkg/vegan/inst/ChangeLog
   pkg/vegan/man/decostand.Rd
Log:
Squashed commit of the following:

commit 3bb2f744f63f96d6a9f2160f8c967749c80e42ab
Merge: fa4e49c cc0ba26
Author: Jari Oksanen <jari.oksanen at oulu.fi>
Date:   Wed May 2 08:49:26 2012 +0300

    Merge branch 'master' of https://github.com/jarioksa/vegan

commit fa4e49ce75abf027347db8d6eca03d3415b0622e
Author: Jari Oksanen <jari.oksanen at oulu.fi>
Date:   Sun Apr 29 09:04:57 2012 +0300

    monoMDS checks that there are enough dissimilarities for the analysis

Modified: pkg/vegan/R/monoMDS.R
===================================================================
--- pkg/vegan/R/monoMDS.R	2012-04-24 18:18:21 UTC (rev 2149)
+++ pkg/vegan/R/monoMDS.R	2012-05-02 05:55:40 UTC (rev 2150)
@@ -91,6 +91,11 @@
     }
     ## ndis: number dissimilarities
     ndis <- length(dist)
+    ## some people try to use NMDS with too few points and too many
+    ## dimensions
+    if (ndis <= k * nobj)
+        stop(gettextf("Not enough data for NMDS: you have %d dissimilarities,\n  and you ask %d scores (%d dimensions times %d points)",
+                      ndis, k * nobj, k, nobj))
     ## starting configuration
     if (missing(y)) {
         y <- matrix(runif(nobj*k, -1, 1), nobj, k)

Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog	2012-04-24 18:18:21 UTC (rev 2149)
+++ pkg/vegan/inst/ChangeLog	2012-05-02 05:55:40 UTC (rev 2150)
@@ -6,6 +6,15 @@
 
 	* rrarefy, drarefy: check the data are integers -- the functions
 	do not give sensible results with real values.
+
+	* monoMDS: checks now that there is a sufficient number of non-NA
+	dissimilarities for the analysis. People really try to use NMDS
+	with too small data sets. The change was triggered by a user who
+	had tried to find a six-dimensional solution for seven points
+	(21 dissimilarities, 42 scores) using 100,000 random starts in
+	metaMDS. With over-defined models there is an infinite number of
+	different solutions with nearly zero stress, and no convergence is
+	found. 
 	
 Version 2.1-14 (opened March 9, 2012)
 

Modified: pkg/vegan/man/decostand.Rd
===================================================================
--- pkg/vegan/man/decostand.Rd	2012-04-24 18:18:21 UTC (rev 2149)
+++ pkg/vegan/man/decostand.Rd	2012-05-02 05:55:40 UTC (rev 2150)
@@ -117,6 +117,10 @@
   estimation of population parameters. \emph{Biometrika}
   \strong{40}, 237--264.
 
+  Good, I. J. 1953. The population frequencies of species and the
+  estimation of population parameters. \emph{Biometrika}
+  \strong{40}, 237--264.
+
   Legendre, P. & Gallagher, E.D. (2001) Ecologically meaningful
   transformations for ordination of species data. \emph{Oecologia}
   \strong{129}, 271--280.