[Vegan-commits] r2710 - in pkg/vegan: inst tests
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Fri Nov 15 14:15:07 CET 2013
Author: jarioksa
Date: 2013-11-15 14:15:07 +0100 (Fri, 15 Nov 2013)
New Revision: 2710
Modified:
pkg/vegan/inst/ChangeLog
pkg/vegan/tests/vegan-tests.Rout.save
Log:
update vegan-tests
Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog 2013-11-15 13:02:44 UTC (rev 2709)
+++ pkg/vegan/inst/ChangeLog 2013-11-15 13:15:07 UTC (rev 2710)
@@ -15,6 +15,16 @@
* oecosimu: change printed quantiles to match the direction of the
test as changed in r2495.
+
+ * tests: vegan examples and vegan-tests have been out of sync for
+ a long time. These have not been updated because most of the
+ changes seem to be triggered by switching to R 3.0-x, and we have
+ not had time to analyse the reasons. The differences also seem to
+ be platform specific, and Linux and MacOS give slightly different
+ results. In particular, there seem to be differences in
+ permutations, constrained ordination, in particular in capscale()
+ and rounding of output. We have also introduced some changes in
+ output that were not yet synced.
Version 2.1-38 (closed November 10, 2013)
Modified: pkg/vegan/tests/vegan-tests.Rout.save
===================================================================
--- pkg/vegan/tests/vegan-tests.Rout.save 2013-11-15 13:02:44 UTC (rev 2709)
+++ pkg/vegan/tests/vegan-tests.Rout.save 2013-11-15 13:15:07 UTC (rev 2710)
@@ -1,8 +1,7 @@
-R version 2.15.3 (2013-03-01) -- "Security Blanket"
+R Under development (unstable) (2013-11-15 r64218) -- "Unsuffered Consequences"
Copyright (C) 2013 The R Foundation for Statistical Computing
-ISBN 3-900051-07-0
-Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
+Platform: x86_64-unknown-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
@@ -63,7 +62,7 @@
Model 6 1.3178 1.3341 99 0.07 .
Residual 4 0.6585
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(m, by="term", perm=100)
Permutation test for cca under reduced model
Terms added sequentially (first to last)
@@ -75,7 +74,7 @@
spno 1 0.1558 0.9461 99 0.40
Residual 4 0.6585
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(m, by="margin", perm=100)
Permutation test for cca under reduced model
Marginal effects of terms
@@ -97,7 +96,7 @@
CCA6 1 0.0681 0.4139 99 0.83
Residual 4 0.6585
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> ## capscale
> p <- capscale(fla, data=df, na.action=na.exclude, subset = Use != "Pasture" & spno > 7)
> anova(p, perm=100)
@@ -108,7 +107,7 @@
Model 6 64.324 1.9652 99 0.01 **
Residual 4 21.821
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(p, by="term", perm=100)
Permutation test for capscale under reduced model
Terms added sequentially (first to last)
@@ -120,7 +119,7 @@
spno 1 7.462 1.3679 99 0.19
Residual 4 21.821
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(p, by="margin", perm=100)
Permutation test for capscale under reduced model
Marginal effects of terms
@@ -132,7 +131,7 @@
spno 1 7.462 1.3679 99 0.31
Residual 4 21.821
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(p, by="axis", perm=100)
Model: capscale(formula = dune ~ Management + poly(A1, 2) + spno, data = df, na.action = na.exclude, subset = structure(c(TRUE, TRUE, TRUE, FALSE, TRUE, FALSE, FALSE, TRUE, FALSE, TRUE, TRUE, FALSE, TRUE, TRUE, TRUE, TRUE, FALSE, TRUE, TRUE, FALSE), .Names = c("2", "13", "4", "16", "6", "1", "8", "5", "17", "15", "10", "11", "9", "18", "3", "20", "14", "19", "12", "7")))
Df Var F N.Perm Pr(>F)
@@ -144,7 +143,7 @@
CAP6 1 2.7811 0.5098 99 0.87
Residual 4 21.8213
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> ## see that capscale can be updated and also works with 'dist' input
> dis <- vegdist(dune)
> p <- update(p, dis ~ .)
@@ -156,7 +155,7 @@
Model 6 1.55041 1.9024 99 0.06 .
Residual 4 0.54333
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(p, by="term", perm=100)
Permutation test for capscale under reduced model
Terms added sequentially (first to last)
@@ -168,7 +167,7 @@
spno 1 0.20517 1.5105 99 0.21
Residual 4 0.54333
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(p, by="margin", perm=100)
Permutation test for capscale under reduced model
Marginal effects of terms
@@ -190,7 +189,7 @@
CAP6 1 0.02458 0.1810 99 0.98
Residual 4 0.54333
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> ### attach()ed data frame instead of data=
> attach(df)
> q <- cca(fla, na.action = na.omit, subset = Use != "Pasture" & spno > 7)
@@ -212,7 +211,7 @@
spno 1 0.1558 0.9461 99 0.43
Residual 4 0.6585
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> anova(q, by="margin", perm=100)
Permutation test for cca under reduced model
Marginal effects of terms
@@ -234,7 +233,7 @@
CCA6 1 0.0681 0.4139 99 0.75
Residual 4 0.6585
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> ### Check that constrained ordination functions can be embedded.
> ### The data.frame 'df' is still attach()ed.
> foo <- function(bar, Y, X, ...)
@@ -266,9 +265,9 @@
Call: rda(formula = Y ~ X, na.action = ..1)
Inertia Proportion Rank
-Total 81.8333 1.0000
-Constrained 28.0931 0.3433 3
-Unconstrained 53.7402 0.6567 14
+Total 81.8300 1.0000
+Constrained 28.0900 0.3433 3
+Unconstrained 53.7400 0.6567 14
Inertia is variance
2 observations deleted due to missingness
@@ -286,9 +285,9 @@
Call: bar(formula = Y ~ X, distance = "jaccard", na.action = ..2)
Inertia Proportion Rank
-Total 5.2931 1.0000
+Total 5.2930 1.0000
Constrained 1.5460 0.2921 3
-Unconstrained 3.7471 0.7079 14
+Unconstrained 3.7470 0.7079 14
Inertia is squared Jaccard distance
2 observations deleted due to missingness
@@ -306,11 +305,11 @@
Call: bar(formula = Y ~ X, na.action = ..1)
Inertia Proportion Rank
-Total 3.9491
-Real Total 4.1689 1.0000
-Constrained 1.3488 0.3235 3
-Unconstrained 2.8201 0.6765 12
-Imaginary -0.2198 5
+Total 3.9490
+Real Total 4.1690 1.0000
+Constrained 1.3490 0.3235 3
+Unconstrained 2.8200 0.6765 12
+Imaginary -0.2200 5
Inertia is squared Bray distance
2 observations deleted due to missingness
@@ -358,7 +357,7 @@
CCA4 1 0.0562 0.6132 99 0.75
Residual 10 0.9170
---
-Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
> all.equal(tab[,2], c(m$CCA$eig, m$CA$tot.chi), check.attributes=FALSE)
[1] TRUE
> tab[nrow(tab),1] == m$CA$rank
@@ -384,9 +383,9 @@
> anova(cap.model.cond, by="axis", strata=CC) # -> error pre r2287
Model: capscale(formula = X ~ A + B + Condition(CC))
Df Var F N.Perm Pr(>F)
-CAP1 1 0.2682 1.3075 99 0.29
-CAP2 1 0.0685 0.3339 99 0.91
-CAP3 1 0.0455 0.2217 99 0.97
+CAP1 1 0.2682 1.3075 99 0.22
+CAP2 1 0.0685 0.3339 99 0.95
+CAP3 1 0.0455 0.2217 99 0.94
Residual 22 4.5130
> anova(cap.model.cond, by="terms", strata=CC) # -> error pre r2287
Permutation test for capscale under reduced model
@@ -395,7 +394,7 @@
Model: capscale(formula = X ~ A + B + Condition(CC))
Df Var F N.Perm Pr(>F)
-A 1 0.1316 0.6415 99 0.72
+A 1 0.1316 0.6415 99 0.71
B 2 0.2506 0.6108 99 0.84
Residual 22 4.5130
>
@@ -404,9 +403,9 @@
> anova(cap.model, by="axis", strata=CC) # -> no error
Model: capscale(formula = X ~ A + B)
Df Var F N.Perm Pr(>F)
-CAP1 1 0.2682 1.3267 99 0.18
-CAP2 1 0.0685 0.3388 99 0.96
-CAP3 1 0.0455 0.2249 99 0.95
+CAP1 1 0.2682 1.3267 99 0.25
+CAP2 1 0.0685 0.3388 99 0.95
+CAP3 1 0.0455 0.2249 99 0.98
Residual 26 5.2565
> anova(cap.model, by="terms", strata=CC) # -> no error
Permutation test for capscale under reduced model
@@ -415,8 +414,8 @@
Model: capscale(formula = X ~ A + B)
Df Var F N.Perm Pr(>F)
-A 1 0.1316 0.6509 99 0.62
-B 2 0.2506 0.6198 99 0.85
+A 1 0.1316 0.6509 99 0.65
+B 2 0.2506 0.6198 99 0.84
Residual 26 5.2565
>
> # partial RDA
@@ -424,9 +423,9 @@
> anova(rda.model.cond, by="axis", strata=CC) # -> no error
Model: rda(formula = X ~ A + B + Condition(CC))
Df Var F N.Perm Pr(>F)
-RDA1 1 0.2682 1.3075 99 0.28
-RDA2 1 0.0685 0.3339 99 0.92
-RDA3 1 0.0455 0.2217 99 0.97
+RDA1 1 0.2682 1.3075 99 0.31
+RDA2 1 0.0685 0.3339 99 0.85
+RDA3 1 0.0455 0.2217 99 0.98
Residual 22 4.5130
> anova(rda.model.cond, by="terms", strata=CC) # -> error pre r2287
Permutation test for rda under reduced model
@@ -435,8 +434,8 @@
Model: rda(formula = X ~ A + B + Condition(CC))
Df Var F N.Perm Pr(>F)
-A 1 0.1316 0.6415 99 0.65
-B 2 0.2506 0.6108 99 0.84
+A 1 0.1316 0.6415 99 0.63
+B 2 0.2506 0.6108 99 0.80
Residual 22 4.5130
>
> # RDA without conditional factor
@@ -444,9 +443,9 @@
> anova(rda.model, by="axis", strata=CC) # -> no error
Model: rda(formula = X ~ A + B)
Df Var F N.Perm Pr(>F)
-RDA1 1 0.2682 1.3267 99 0.24
-RDA2 1 0.0685 0.3388 99 0.90
-RDA3 1 0.0455 0.2249 99 0.94
+RDA1 1 0.2682 1.3267 99 0.21
+RDA2 1 0.0685 0.3388 99 0.83
+RDA3 1 0.0455 0.2249 99 0.97
Residual 26 5.2565
> anova(rda.model, by="terms", strata=CC) # -> no error
Permutation test for rda under reduced model
@@ -455,8 +454,8 @@
Model: rda(formula = X ~ A + B)
Df Var F N.Perm Pr(>F)
-A 1 0.1316 0.6509 99 0.69
-B 2 0.2506 0.6198 99 0.84
+A 1 0.1316 0.6509 99 0.77
+B 2 0.2506 0.6198 99 0.83
Residual 26 5.2565
> ## clean.up
> rm(X, A, B, CC, cap.model.cond, cap.model, rda.model.cond, rda.model)
@@ -511,23 +510,39 @@
> ### end envfit & plot.envfit
>
> ### protest (& Procrustes analysis): Stability of the permutations and
-> ### other results.
+> ### other results.
> data(mite)
> mod <- rda(mite)
> x <- scores(mod, display = "si", choices=1:6)
> set.seed(4711)
> xp <- x[sample(nrow(x)),]
-> pro <- protest(x, xp, permutations = 99)
+> pro <- protest(x, xp, control = how(nperm = 99))
> pro
Call:
-protest(X = x, Y = xp, permutations = 99)
+protest(X = x, Y = xp, control = how(nperm = 99))
Procrustes Sum of Squares (m12 squared): 0.9147
Correlation in a symmetric Procrustes rotation: 0.292
Significance: 0.08
-Based on 99 permutations.
+Based on 99 permutations
+Permutation Design:
+Blocks:
+ Defined by: none
+
+Plots:
+ Defined by: none
+
+Within Plots:
+ Permutation type: free
+
+Permutation details:
+ Number of permutations requested: 99
+ Max. number of permutations allowed: 9999
+ Evaluate all permutations?: No. Activation limit: 99
+.
+
> pro$t
[1] 0.2056649 0.2008622 0.2062525 0.2656158 0.2739148 0.1990895 0.3124461
[8] 0.2287779 0.2033140 0.2396172 0.1868019 0.2736072 0.1830199 0.2026019
@@ -579,4 +594,4 @@
>
> proc.time()
user system elapsed
- 9.182 0.136 9.543
+ 2.387 0.039 2.415
More information about the Vegan-commits
mailing list