[Vegan-commits] r2626 - in pkg/vegan: R inst

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

Author: jarioksa
Date: 2013-09-26 12:27:31 +0200 (Thu, 26 Sep 2013)
New Revision: 2626

Modified:
   pkg/vegan/R/nestednodf.R
   pkg/vegan/inst/ChangeLog
Log:
nestednodf(..., weighted=TRUE) gave wrong fill and was inconsistent if data were binary

Modified: pkg/vegan/R/nestednodf.R
===================================================================
--- pkg/vegan/R/nestednodf.R	2013-09-26 10:11:52 UTC (rev 2625)
+++ pkg/vegan/R/nestednodf.R	2013-09-26 10:27:31 UTC (rev 2626)
@@ -22,7 +22,7 @@
     }
     nr <- NROW(comm)
     nc <- NCOL(comm)
-    fill <- sum(rfill)/length(comm)
+    fill <- sum(rfill)/prod(dim(comm))
     N.paired.rows <- numeric(nr * (nr - 1)/2)
     N.paired.cols <- numeric(nc * (nc - 1)/2)
     counter <- 0
@@ -35,7 +35,7 @@
             if (weighted) {
                 second <- comm[j, ]
                 N.paired.rows[counter] <-
-                    sum(first - second > 0 & second > 0)/sum(second > 0)
+                    sum(first - second >= 0 & second > 0)/sum(second > 0)
             }
             else {
                 N.paired.rows[counter] <-
@@ -53,7 +53,7 @@
             if (weighted) {
                 second <- comm[, j]
                 N.paired.cols[counter] <-
-                    sum(first - second > 0 & second > 0)/sum(second > 0)
+                    sum(first - second >= 0 & second > 0)/sum(second > 0)
             }
             else {
                 N.paired.cols[counter] <-

Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog	2013-09-26 10:11:52 UTC (rev 2625)
+++ pkg/vegan/inst/ChangeLog	2013-09-26 10:27:31 UTC (rev 2626)
@@ -6,6 +6,20 @@
 
 	* opened with the release of vegan 2.0-9.
 
+	* nestednodf: matrix fill was wrongly calculated in weighted
+	analysis. The nominator was length of 'comm', and if input was a
+	data frame that was the number of columns instead of the number of
+	cells. The fill was correct in non-weighted analysis because there
+	data were transformed to a matrix, and the length of a matrix is
+	the number of cells (unlike in data frames).
+
+	* nestednodf: weighted analysis gave all statistics as zero if
+	binary data were supplied. Some ">" comparisons were changed to
+	">=" and now weighted analysis of binary data gives same results
+	as non-weighted analysis. However, this can change results of
+	weighted analysis of quantitative data. The change needs
+	endorsement by the function author Gustavo Carvalho.
+
 	* oecosimu: warns user if the specified nullmodel 'method' changes
 	quantitative input data to binary data under cover. Some people
 	have not noticed this.