[Vegan-commits] r1318 - in pkg/vegan: . R inst

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

Author: jarioksa
Date: 2010-09-26 16:45:07 +0200 (Sun, 26 Sep 2010)
New Revision: 1318

Modified:
   pkg/vegan/DESCRIPTION
   pkg/vegan/R/decorana.R
   pkg/vegan/inst/ChangeLog
Log:
fix true eigenvalues in decorana

Modified: pkg/vegan/DESCRIPTION
===================================================================
--- pkg/vegan/DESCRIPTION	2010-09-24 12:18:17 UTC (rev 1317)
+++ pkg/vegan/DESCRIPTION	2010-09-26 14:45:07 UTC (rev 1318)
@@ -1,7 +1,7 @@
 Package: vegan
 Title: Community Ecology Package
-Version: 1.18-12
-Date: September 17, 2010
+Version: 1.18-13
+Date: September 26, 2010
 Author: Jari Oksanen, F. Guillaume Blanchet, Roeland Kindt, Pierre Legendre, 
    R. B. O'Hara, Gavin L. Simpson, Peter Solymos, M. Henry H. Stevens, 
    Helene Wagner  

Modified: pkg/vegan/R/decorana.R
===================================================================
--- pkg/vegan/R/decorana.R	2010-09-24 12:18:17 UTC (rev 1317)
+++ pkg/vegan/R/decorana.R	2010-09-26 14:45:07 UTC (rev 1318)
@@ -141,11 +141,7 @@
     }
     else {
         evals.decorana <- evals
-        var.r <- cov.wt(rproj, aidot)
-        var.r <- diag(var.r$cov) * (1 - sum(var.r$wt^2))
-        var.c <- cov.wt(cproj, adotj)
-        var.c <- diag(var.c$cov) * (1 - sum(var.c$wt^2))
-        evals <- var.r/var.c
+        evals <- eigengrad(rproj, veg)
     }
     CA <- list(rproj = rproj, cproj = cproj, evals = evals, evals.decorana = evals.decorana, 
                origin = origin, v = v, fraction = v.fraction, adotj = adotj, 

Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog	2010-09-24 12:18:17 UTC (rev 1317)
+++ pkg/vegan/inst/ChangeLog	2010-09-26 14:45:07 UTC (rev 1318)
@@ -2,8 +2,14 @@
 
 VEGAN DEVEL VERSIONS at http://r-forge.r-project.org/
 
-Version 1.18-12 (opened September 17, 2010)
+Version 1.18-13 (opened September 26, 2010)
 
+	* decorana: fixed true eigenvalues which were slightly off (too
+	low). Now uses eigengrad() which is more careful in calculations
+	(like in centring).  
+	
+Version 1.18-12 (closed September 26, 2010)
+
 	* swan: no eternal loop with disconnected data where zeros will
 	remain in the matrix.