[Vegan-commits] r2034 - branches/2.0/R branches/2.0/inst branches/2.0/man www
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Wed Dec 21 13:14:56 CET 2011
Author: jarioksa
Date: 2011-12-21 13:14:55 +0100 (Wed, 21 Dec 2011)
New Revision: 2034
Modified:
branches/2.0/R/oecosimu.R
branches/2.0/R/print.oecosimu.R
branches/2.0/inst/ChangeLog
branches/2.0/inst/NEWS.Rd
branches/2.0/man/oecosimu.Rd
www/NEWS.html
Log:
merge r2027,28,33: tweaks in oecosimu output
Modified: branches/2.0/R/oecosimu.R
===================================================================
--- branches/2.0/R/oecosimu.R 2011-12-21 12:02:21 UTC (rev 2033)
+++ branches/2.0/R/oecosimu.R 2011-12-21 12:14:55 UTC (rev 2034)
@@ -98,7 +98,8 @@
}
## end of addition
sd <- apply(simind, 1, sd, na.rm = TRUE)
- z <- (indstat - rowMeans(simind, na.rm = TRUE))/sd
+ means <- rowMeans(simind, na.rm = TRUE)
+ z <- (indstat - means)/sd
if (any(sd < sqrt(.Machine$double.eps)))
z[sd < sqrt(.Machine$double.eps)] <- 0
pless <- rowSums(indstat <= simind, na.rm = TRUE)
@@ -124,7 +125,8 @@
ind <- list(statistic = ind)
if (method == "custom")
attr(method, "permfun") <- permfun
- ind$oecosimu <- list(z = z, pval = p, simulated=simind, method=method,
+ ind$oecosimu <- list(z = z, means = means, pval = p, simulated=simind,
+ method=method,
statistic = indstat, alternative = alternative)
class(ind) <- c("oecosimu", class(ind))
ind
Modified: branches/2.0/R/print.oecosimu.R
===================================================================
--- branches/2.0/R/print.oecosimu.R 2011-12-21 12:02:21 UTC (rev 2033)
+++ branches/2.0/R/print.oecosimu.R 2011-12-21 12:14:55 UTC (rev 2034)
@@ -12,7 +12,7 @@
two.sided = "not equal to",
less = "less than",
greater = "greater than")
- cat("\nalternative hypothesis: true mean is", alt.char, "the statistic")
+ cat("\nalternative hypothesis: simulated median is", alt.char, "the statistic")
## dim attribute is always there, but print all others
cat("\n\n")
@@ -27,9 +27,9 @@
greater = c(0.05, 0.5, 1))
qu <- apply(x$oecosimu$simulated, 1, quantile, probs=probs, na.rm = TRUE)
m <- cbind("statistic" = x$oecosimu$statistic,
- "z" = x$oecosimu$z, t(qu),
+ "z" = x$oecosimu$z, "mean" = x$oecosimu$means, t(qu),
"Pr(sim.)"=x$oecosimu$pval)
- printCoefmat(m, ...)
+ printCoefmat(m, cs.ind = 3:6, ...)
if (any(is.na(x$oecosimu$simulated))) {
nacount <- rowSums(is.na(x$oecosimu$simulated))
cat("\nNumber of NA cases removed from simulations:\n",
Modified: branches/2.0/inst/ChangeLog
===================================================================
--- branches/2.0/inst/ChangeLog 2011-12-21 12:02:21 UTC (rev 2033)
+++ branches/2.0/inst/ChangeLog 2011-12-21 12:14:55 UTC (rev 2034)
@@ -5,6 +5,7 @@
Version 2.0-3 (opened November 13, 2011)
* merge r2030: resurrect old ChangeLog entries.
+ * merge r2027,28,33: tweak oecosimu printed output.
* merge r2018,19,21: add nestedbetasor() & nestedbetajac().
* merge r2016: empty factor levels in betadisper() fixed in
ordimedian().
Modified: branches/2.0/inst/NEWS.Rd
===================================================================
--- branches/2.0/inst/NEWS.Rd 2011-12-21 12:02:21 UTC (rev 2033)
+++ branches/2.0/inst/NEWS.Rd 2011-12-21 12:14:55 UTC (rev 2034)
@@ -9,8 +9,9 @@
\item Added new nestedness functions \code{nestedbetasor} and
\code{nestedbetajac} that implement multiple-site dissimilarity
- indices of Baselga (\emph{Global Ecology and Biogeography} 19,
- 134--143; 2010).
+ indices and their decomposition into turnover and nestedness
+ components following Baselga (\emph{Global Ecology and
+ Biogeography} 19, 134--143; 2010).
\item Added function \code{rarecurve} to draw rarefaction curves
for each row (sampling unit) of the input data, optionally with
@@ -37,6 +38,10 @@
dissimilarities (\code{stepacross}), instead of being treated
as synonymous with \code{noshare = FALSE} which always
suppresses extended dissimilarities.
+
+ \item \code{oecosimu} displays the mean of simulations and
+ describes alternative hypothesis more clearly in the printed
+ output.
}
} % end new features
Modified: branches/2.0/man/oecosimu.Rd
===================================================================
--- branches/2.0/man/oecosimu.Rd 2011-12-21 12:02:21 UTC (rev 2033)
+++ branches/2.0/man/oecosimu.Rd 2011-12-21 12:14:55 UTC (rev 2034)
@@ -49,7 +49,10 @@
\code{nestedfun}}
\item{alternative}{a character string specifying the alternative
hypothesis, must be one of \code{"two.sided"} (default), \code{"greater"}
- or \code{"less"}. You can specify just the initial letter.}
+ or \code{"less"}. Please note that the \eqn{p}-value of two-sided
+ test is approximately two times higher than in the corresponding
+ one-sided test (\code{"greater"} or \code{"less"} depending on the
+ sign of the difference).}
\item{data}{Ignored argument of the generic function.}
\item{xlab}{Label of the x-axis.}
\item{\dots}{Other arguments to functions.}
@@ -179,14 +182,15 @@
can be dangerous.
}
-\value{
- Function \code{oecosimu} returns the result of \code{nestfun}
- with one added component called \code{oecosimu}. The \code{oecosimu}
+\value{
+ Function \code{oecosimu} returns the result of \code{nestfun} added
+ with a component called \code{oecosimu}. The \code{oecosimu}
component contains the simulated values of the statistic (item
- \code{simulated}), the name of the \code{method}, two-sided \eqn{P}
- value and z-value of the statistic based on simulation. The
- \code{commsimulator} returns a null model matrix or a swap of the
- input matrix. }
+ \code{simulated}), the name of the \code{method}, \eqn{P} value
+ (with given \code{alternative}), \eqn{z}-value of the statistic
+ based on simulation (also known as standardized effect size), and
+ the mean of simulations.
+}
\references{
Gotelli, N.J. & Entsminger, N.J. (2001). Swap and fill algorithms in
Modified: www/NEWS.html
===================================================================
--- www/NEWS.html 2011-12-21 12:02:21 UTC (rev 2033)
+++ www/NEWS.html 2011-12-21 12:14:55 UTC (rev 2034)
@@ -10,70 +10,99 @@
<h3>Changes in version 2.0-3</h3>
+<p>
+</p>
+<p>
-
<h4>NEW FUNCTIONS</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li><p> Added new nestedness functions <code>nestedbetasor</code> and
+<li> Added new nestedness functions <code>nestedbetasor</code> and
<code>nestedbetajac</code> that implement multiple-site dissimilarity
-indices of Baselga (<EM>Global Ecology and Biogeography</EM> 19,
-134–143; 2010).
+indices and their decomposition into turnover and nestedness
+components following Baselga (<EM>Global Ecology and
+Biogeography</EM> 19, 134–143; 2010).
</p>
+<p>
</li>
-<li><p> Added function <code>rarecurve</code> to draw rarefaction curves
+<li>Added function <code>rarecurve</code> to draw rarefaction curves
for each row (sampling unit) of the input data, optionally with
lines showing rarefied species richness with given sample size
for each curve.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>BUG FIXES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>betadisper()</code> failed when the <code>groups</code> was a
+<li> <code>betadisper()</code> failed when the <code>groups</code> was a
factor with empty levels.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>NEW FEATURES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>metaMDS</code> argument <code>noshare = 0</code> is now
+<li> <code>metaMDS</code> argument <code>noshare = 0</code> is now
regarded as a numeric threshold that always triggers extended
dissimilarities (<code>stepacross</code>), instead of being treated
as synonymous with <code>noshare = FALSE</code> which always
suppresses extended dissimilarities.
</p>
+<p>
+</li>
+<li><code>oecosimu</code> displays the mean of simulations and
+describes alternative hypothesis more clearly in the printed
+output.
+</p>
+<p>
</ul>
+</p>
+</p>
<h3>Changes in version 2.0-2</h3>
+<p>
-
<h4>BUG FIXES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li><p> Function <code>capscale</code> failed if constrained component
+<li> Function <code>capscale</code> failed if constrained component
had zero rank. This happened most likely in partial models when
the conditions aliased constraints. The problem was observed in
<code>anova(..., by ="margin")</code> which uses partial models to
@@ -82,8 +111,9 @@
<a href="https://stat.ethz.ch/pipermail/r-help/2011-October/293077.html">R-News
mailing list</a>.
</p>
+<p>
</li>
-<li> <p><code>stressplot</code> and <code>goodness</code> sometimes failed when
+<li><code>stressplot</code> and <code>goodness</code> sometimes failed when
<code>metaMDS</code> was based on <code>isoMDS</code> (<span class="pkg">MASS</span> package)
because <code>metaMDSdist</code> did not use the same defaults for
step-across (extended) dissimilarities as <code>metaMDS(...,
@@ -91,8 +121,9 @@
triggering of step-across in <code>capscale(..., metaMDSdist =
TRUE)</code>.
</p>
+<p>
</li>
-<li> <p><code>adonis</code> contained a minor bug resulting from
+<li><code>adonis</code> contained a minor bug resulting from
incomplete implementation of a speed-up that did not affect the
results. In fixing this bug, a further bug was identified in
transposing the hat matrices. This second bug was only active
@@ -100,57 +131,73 @@
speed-up in the internal f.test() function is fully
realised. Reported by Nicholas Lewin-Koh.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>NEW FEATURES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>ordiarrows</code> and <code>ordisegments</code> gained argument
+<li> <code>ordiarrows</code> and <code>ordisegments</code> gained argument
<code>order.by</code> that gives a variable to sort points within
<code>groups</code>. Earlier the points were assumed to be in order.
</p>
+<p>
</li>
-<li><p> Function <code>ordispider</code> invisibly returns the
+<li>Function <code>ordispider</code> invisibly returns the
coordinates to which the points were connected. Typically these
are class centroids of each point, but for constrained ordination
with no <code>groups</code> they are the LC scores.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
+</p>
<h3>Changes in version 2.0-1</h3>
+<p>
-
<h4>NEW FUNCTIONS</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>clamtest</code>: new function to classify species as
+<li> <code>clamtest</code>: new function to classify species as
generalists and specialists in two distinct habitats (CLAM test of
Chazdon et al., <EM>Ecology</EM> 92, 1332–1343; 2011). The test is
based on multinomial distribution of individuals in two habitat
types or sampling units, and it is applicable only to count data
with no over-dispersion.
</p>
+<p>
</li>
-<li> <p><code>as.preston</code> gained <code>plot</code> and <code>lines</code>
+<li><code>as.preston</code> gained <code>plot</code> and <code>lines</code>
methods, and <code>as.fisher</code> gained <code>plot</code> method (which
also can add items to existing plots). These are similar as
<code>plot</code> and <code>lines</code> for <code>prestonfit</code> and
<code>fisherfit</code>, but display only data without the fitted lines.
</p>
+<p>
</li>
-<li> <p><code>raupcrick</code>: new function to implement Raup-Crick
+<li><code>raupcrick</code>: new function to implement Raup-Crick
dissimilarity as a probability of number of co-occurring species
with occurrence probabilities proportional to species
frequencies. <span class="pkg">Vegan</span> has Raup-Crick index as a choice in
@@ -161,37 +208,47 @@
[<a href="http://www.esajournals.org/doi/abs/10.1890/ES10-00117.1">doi:10.1890/ES10-00117.1</a>],
and was developed with the consultation of Brian Inouye.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>BUG FIXES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li><p> Function <code>meandist</code> could scramble items and give
+<li> Function <code>meandist</code> could scramble items and give
wrong results, especially when the <code>grouping</code> was
numerical. The problem was reported by Dr Miguel Alvarez
(Univ. Bonn).
</p>
+<p>
</li>
-<li> <p><code>metaMDS</code> did not reset <code>tries</code> when a new model
+<li><code>metaMDS</code> did not reset <code>tries</code> when a new model
was started with a <code>previous.best</code> solution from a different
model.
</p>
+<p>
</li>
-<li><p> Function <code>permatswap</code> for community null models using
+<li>Function <code>permatswap</code> for community null models using
quantitative swap never swapped items in a <i>2 by 2</i>
submatrix if all cells were filled.
</p>
+<p>
</li>
-<li><p> The result from <code>permutest.cca</code> could not be
+<li>The result from <code>permutest.cca</code> could not be
<code>update</code>d because of a ‘<span class="file">NAMESPACE</span>’ issue.
</p>
+<p>
</li>
-<li> <p><font face="Courier New,Courier" color="#666666"><b>R</b></font> 2.14.0 changed so that it does not accept using
+<li><font face="Courier New,Courier" color="#666666"><b>R</b></font> 2.14.0 changed so that it does not accept using
<code>sd()</code> function for matrices (which was the behaviour at
least since <font face="Courier New,Courier" color="#666666"><b>R</b></font> 1.0-0), and several <span class="pkg">vegan</span> functions were
changed to adapt to this change (<code>rda</code>, <code>capscale</code>,
@@ -200,40 +257,54 @@
results but you probably wish to upgrade <span class="pkg">vegan</span> to avoid
annoying warnings.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>ANALYSES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>nesteddisc</code> is slacker and hence faster when trying
+<li> <code>nesteddisc</code> is slacker and hence faster when trying
to optimize the statistic for tied column frequencies. Tracing
showed that in most cases an improved ordering was found rather
early in tries, and the results are equally good in most cases.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
+</p>
<h3>Changes in version 2.0-0</h3>
+<p>
-
<h4>GENERAL</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li><p> Peter Minchin joins the <span class="pkg">vegan</span> team.
+<li> Peter Minchin joins the <span class="pkg">vegan</span> team.
</p>
+<p>
</li>
-<li> <p><span class="pkg">vegan</span> implements standard <font face="Courier New,Courier" color="#666666"><b>R</b></font> ‘<span class="file">NAMESPACE</span>’. In
+<li><span class="pkg">vegan</span> implements standard <font face="Courier New,Courier" color="#666666"><b>R</b></font> ‘<span class="file">NAMESPACE</span>’. In
general, <code>S3</code> methods are not exported which means that you
cannot directly use or see contents of functions like
<code>cca.default</code>, <code>plot.cca</code> or <code>anova.ccabyterm</code>. To
@@ -244,24 +315,31 @@
<code>vegan:::cca.default</code>. This change may break packages,
documents or scripts that rely on non-exported names.
</p>
+<p>
</li>
-<li> <p><span class="pkg">vegan</span> depends on the <span class="pkg">permute</span> package. This
+<li><span class="pkg">vegan</span> depends on the <span class="pkg">permute</span> package. This
package provides powerful tools for restricted permutation
schemes. All <span class="pkg">vegan</span> permutation will gradually move to use
<span class="pkg">permute</span>, but currently only <code>betadisper</code> uses the new
feature.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>NEW FUNCTIONS</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>monoMDS</code>: a new function for non-metric
+<li> <code>monoMDS</code>: a new function for non-metric
multidimensional scaling (NMDS). This function replaces
<code>MASS::isoMDS</code> as the default method in
<code>metaMDS</code>. Major advantages of <code>monoMDS</code> are that it
@@ -275,22 +353,25 @@
previous alternatives. The function uses <code>Fortran</code> code
written by Peter Minchin.
</p>
+<p>
</li>
-<li> <p><code>MDSrotate</code> a new function to replace
+<li><code>MDSrotate</code> a new function to replace
<code>metaMDSrotate</code>. This function can rotate both <code>metaMDS</code>
and <code>monoMDS</code> results so that the first axis is parallel to
an environmental vector.
</p>
+<p>
</li>
-<li> <p><code>eventstar</code> finds the minimum of the evenness profile
+<li><code>eventstar</code> finds the minimum of the evenness profile
on the Tsallis entropy, and uses this to find the corresponding
values of diversity, evenness and numbers equivalent following
Mendes et al. (<EM>Ecography</EM> 31, 450-456; 2008). The code was
contributed by Eduardo Ribeira Cunha and Heloisa Beatriz Antoniazi
Evangelista and adapted to <span class="pkg">vegan</span> by Peter Solymos.
</p>
+<p>
</li>
-<li> <p><code>fitspecaccum</code> fits non-linear regression models to
+<li><code>fitspecaccum</code> fits non-linear regression models to
the species accumulation results from <code>specaccum</code>. The
function can use new self-starting species accumulation models
in <span class="pkg">vegan</span> or other self-starting non-linear regression
@@ -299,36 +380,45 @@
Michaelis-Menten, logistic and Weibull (in base <font face="Courier New,Courier" color="#666666"><b>R</b></font>) models. The
function has <code>plot</code> and <code>predict</code> methods.
</p>
+<p>
</li>
-<li><p> Self-starting non-linear species accumulation models
+<li>Self-starting non-linear species accumulation models
<code>SSarrhenius</code>, <code>SSgleason</code>, <code>SSgitay</code> and
<code>SSlomolino</code>. These can be used with <code>fitspecaccum</code> or
directly in non-linear regression with <code>nls</code>. These functions
were implemented because they were found good for species-area
models by Dengler (<EM>J. Biogeogr.</EM> 36, 728-744; 2009).
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>NEW FEATURES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>adonis</code>, <code>anosim</code>, <code>meandist</code> and
+<li> <code>adonis</code>, <code>anosim</code>, <code>meandist</code> and
<code>mrpp</code> warn on negative dissimilarities, and
<code>betadisper</code> refuses to analyse them. All these functions
expect dissimilarities, and giving something else (like
correlations) probably is a user error.
</p>
+<p>
</li>
-<li> <p><code>betadisper</code> uses restricted permutation of the
+<li><code>betadisper</code> uses restricted permutation of the
<span class="pkg">permute</span> package.
</p>
+<p>
</li>
-<li> <p><code>metaMDS</code> uses <code>monoMDS</code> as its default ordination
+<li><code>metaMDS</code> uses <code>monoMDS</code> as its default ordination
engine. Function gains new argument <code>engine</code> that can be used
to alternatively select <code>MASS::isoMDS</code>. The default is not
to use <code>stepacross</code> with <code>monoMDS</code> because its
@@ -337,53 +427,73 @@
with <code>isoMDS</code> because it cannot handle adequately these tied
maximum dissimilarities.
</p>
+<p>
</li>
-<li> <p><code>specaccum</code> gained <code>predict</code> method which uses
+<li><code>specaccum</code> gained <code>predict</code> method which uses
either linear or spline interpolation for data between observed
points. Extrapolation is possible with spline interpolation, but
may make little sense.
</p>
+<p>
</li>
-<li> <p><code>specpool</code> can handle missing values or empty factor
+<li><code>specpool</code> can handle missing values or empty factor
levels in the grouping factor <code>pool</code>. Now also checks that
the length of the <code>pool</code> matches the number of
observations.
</p>
+<p>
</ul>
+</p>
+</p>
+<p>
<h4>DEPRECATED AND DEFUNCT</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>metaMDSrotate</code> was replaced with <code>MDSrotate</code>
+<li> <code>metaMDSrotate</code> was replaced with <code>MDSrotate</code>
that can also handle the results of <code>monoMDS</code>.
</p>
+<p>
</li>
-<li> <p><code>permuted.index2</code> and other “new” permutation
+<li><code>permuted.index2</code> and other “new” permutation
code was removed in favour of the <span class="pkg">permute</span> package. This code
was not intended for normal use, but packages depending on that
code in <span class="pkg">vegan</span> should instead depend on <span class="pkg">permute</span>.
</p>
+<p>
</ul>
+</p>
+<p>
+</p>
+</p>
+<p>
<h4>ANALYSES</h4>
+<p>
-
+</p>
+<p>
<ul>
-<li> <p><code>treeheight</code> uses much snappier code. The results
+<li> <code>treeheight</code> uses much snappier code. The results
should be unchanged.
</p>
+<p>
</ul>
+</p>
+</p>
</body></html>
More information about the Vegan-commits
mailing list