698 lines
39 KiB
HTML
698 lines
39 KiB
HTML
|
<!DOCTYPE html>
|
|||
|
|
<html>
|
|||
|
|
<head>
|
|||
|
|
<meta charset="utf-8">
|
|||
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
|||
|
|
<title><strong>vegan</strong> FAQ</title>
|
|||
|
|
<style type="text/css">
|
|||
|
|
/**
|
|||
|
|
* Prism.s theme ported from highlight.js's xcode style
|
|||
|
|
*/
|
|||
|
|
pre code {
|
|||
|
|
padding: 1em;
|
|||
|
|
}
|
|||
|
|
.token.comment {
|
|||
|
|
color: #007400;
|
|||
|
|
}
|
|||
|
|
.token.punctuation {
|
|||
|
|
color: #999;
|
|||
|
|
}
|
|||
|
|
.token.tag,
|
|||
|
|
.token.selector {
|
|||
|
|
color: #aa0d91;
|
|||
|
|
}
|
|||
|
|
.token.boolean,
|
|||
|
|
.token.number,
|
|||
|
|
.token.constant,
|
|||
|
|
.token.symbol {
|
|||
|
|
color: #1c00cf;
|
|||
|
|
}
|
|||
|
|
.token.property,
|
|||
|
|
.token.attr-name,
|
|||
|
|
.token.string,
|
|||
|
|
.token.char,
|
|||
|
|
.token.builtin {
|
|||
|
|
color: #c41a16;
|
|||
|
|
}
|
|||
|
|
.token.inserted {
|
|||
|
|
background-color: #ccffd8;
|
|||
|
|
}
|
|||
|
|
.token.deleted {
|
|||
|
|
background-color: #ffebe9;
|
|||
|
|
}
|
|||
|
|
.token.operator,
|
|||
|
|
.token.entity,
|
|||
|
|
.token.url,
|
|||
|
|
.language-css .token.string,
|
|||
|
|
.style .token.string {
|
|||
|
|
color: #9a6e3a;
|
|||
|
|
}
|
|||
|
|
.token.atrule,
|
|||
|
|
.token.attr-value,
|
|||
|
|
.token.keyword {
|
|||
|
|
color: #836c28;
|
|||
|
|
}
|
|||
|
|
.token.function,
|
|||
|
|
.token.class-name {
|
|||
|
|
color: #DD4A68;
|
|||
|
|
}
|
|||
|
|
.token.regex,
|
|||
|
|
.token.important,
|
|||
|
|
.token.variable {
|
|||
|
|
color: #5c2699;
|
|||
|
|
}
|
|||
|
|
.token.important,
|
|||
|
|
.token.bold {
|
|||
|
|
font-weight: bold;
|
|||
|
|
}
|
|||
|
|
.token.italic {
|
|||
|
|
font-style: italic;
|
|||
|
|
}
|
|||
|
|
</style>
|
|||
|
|
<style type="text/css">
|
|||
|
|
body {
|
|||
|
|
font-family: sans-serif;
|
|||
|
|
max-width: 800px;
|
|||
|
|
margin: auto;
|
|||
|
|
padding: 1em;
|
|||
|
|
line-height: 1.5;
|
|||
|
|
box-sizing: border-box;
|
|||
|
|
}
|
|||
|
|
body, .footnotes, code { font-size: .9em; }
|
|||
|
|
li li { font-size: .95em; }
|
|||
|
|
*, *:before, *:after {
|
|||
|
|
box-sizing: inherit;
|
|||
|
|
}
|
|||
|
|
pre, img { max-width: 100%; }
|
|||
|
|
pre, pre:hover {
|
|||
|
|
white-space: pre-wrap;
|
|||
|
|
word-break: break-all;
|
|||
|
|
}
|
|||
|
|
pre code {
|
|||
|
|
display: block;
|
|||
|
|
overflow-x: auto;
|
|||
|
|
}
|
|||
|
|
code { font-family: 'DejaVu Sans Mono', 'Droid Sans Mono', 'Lucida Console', Consolas, Monaco, monospace; }
|
|||
|
|
:not(pre) > code, code[class] { background-color: #F8F8F8; }
|
|||
|
|
code.language-undefined, pre > code:not([class]) {
|
|||
|
|
background-color: inherit;
|
|||
|
|
border: 1px solid #eee;
|
|||
|
|
}
|
|||
|
|
table {
|
|||
|
|
margin: auto;
|
|||
|
|
border-top: 1px solid #666;
|
|||
|
|
}
|
|||
|
|
table thead th { border-bottom: 1px solid #ddd; }
|
|||
|
|
th, td { padding: 5px; }
|
|||
|
|
thead, tfoot, tr:nth-child(even) { background: #eee; }
|
|||
|
|
blockquote {
|
|||
|
|
color: #666;
|
|||
|
|
margin: 0;
|
|||
|
|
padding-left: 1em;
|
|||
|
|
border-left: 0.5em solid #eee;
|
|||
|
|
}
|
|||
|
|
hr, .footnotes::before { border: 1px dashed #ddd; }
|
|||
|
|
.frontmatter { text-align: center; }
|
|||
|
|
#TOC .numbered li { list-style: none; }
|
|||
|
|
#TOC .numbered { padding-left: 0; }
|
|||
|
|
#TOC .numbered ul { padding-left: 1em; }
|
|||
|
|
table, .body h2 { border-bottom: 1px solid #666; }
|
|||
|
|
.body .appendix, .appendix ~ h2 { border-bottom-style: dashed; }
|
|||
|
|
.footnote-ref a::before { content: "["; }
|
|||
|
|
.footnote-ref a::after { content: "]"; }
|
|||
|
|
section.footnotes::before {
|
|||
|
|
content: "";
|
|||
|
|
display: block;
|
|||
|
|
max-width: 20em;
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
@media print {
|
|||
|
|
body {
|
|||
|
|
font-size: 12pt;
|
|||
|
|
max-width: 100%;
|
|||
|
|
}
|
|||
|
|
tr, img { page-break-inside: avoid; }
|
|||
|
|
}
|
|||
|
|
@media only screen and (min-width: 992px) {
|
|||
|
|
pre { white-space: pre; }
|
|||
|
|
}
|
|||
|
|
</style>
|
|||
|
|
</head>
|
|||
|
|
<body>
|
|||
|
|
<div class="frontmatter">
|
|||
|
|
<div class="title"><h1><strong>vegan</strong> FAQ</h1></div>
|
|||
|
|
<div class="author"><h2></h2></div>
|
|||
|
|
<div class="date"><h3></h3></div>
|
|||
|
|
</div>
|
|||
|
|
<div class="body">
|
|||
|
|
<!--
|
|||
|
|
%\VignetteIndexEntry{vegan FAQ}
|
|||
|
|
%\VignetteEngine{knitr::knitr}
|
|||
|
|
%\VignetteEncoding{UTF-8}
|
|||
|
|
-->
|
|||
|
|
<h1 id="vegan-faq"><strong>vegan</strong> FAQ</h1>
|
|||
|
|
<p>This document contains answers to some of the most frequently asked
|
|||
|
|
questions about R package <strong>vegan</strong>.</p>
|
|||
|
|
<blockquote>
|
|||
|
|
<p>This work is licensed under the Creative Commons Attribution 3.0
|
|||
|
|
License. To view a copy of this license, visit
|
|||
|
|
<a href="https://creativecommons.org/licenses/by/3.0/">https://creativecommons.org/licenses/by/3.0/</a> or send a letter to
|
|||
|
|
Creative Commons, 543 Howard Street, 5th Floor, San Francisco,
|
|||
|
|
California, 94105, USA.</p>
|
|||
|
|
<p>Copyright © 2008-2016 vegan development team</p>
|
|||
|
|
</blockquote>
|
|||
|
|
<hr />
|
|||
|
|
<h2 id="introduction">Introduction</h2>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-is-vegan">What is <strong>vegan</strong>?</h3>
|
|||
|
|
<p><strong>Vegan</strong> is an R package for community ecologists. It contains the most
|
|||
|
|
popular methods of multivariate analysis needed in analysing ecological
|
|||
|
|
communities, and tools for diversity analysis, and other potentially
|
|||
|
|
useful functions. <strong>Vegan</strong> is not self-contained but it must be run
|
|||
|
|
under R statistical environment, and it also depends on many other R
|
|||
|
|
packages. <strong>Vegan</strong> is <a href="https://www.gnu.org/philosophy/free-sw.html">free
|
|||
|
|
software</a> and distributed
|
|||
|
|
under <a href="https://www.gnu.org/licenses/gpl.html">GPL2 license</a>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-is-r">What is R?</h3>
|
|||
|
|
<p>R is a system for statistical computation and graphics. It consists of a
|
|||
|
|
language plus a run-time environment with graphics, a debugger, access
|
|||
|
|
to certain system functions, and the ability to run programs stored in
|
|||
|
|
script files.</p>
|
|||
|
|
<p>R has a home page at <a href="https://www.R-project.org/">https://www.R-project.org/</a>. It is <a href="https://www.gnu.org/philosophy/free-sw.html">free
|
|||
|
|
software</a> distributed under
|
|||
|
|
a GNU-style <a href="https://www.gnu.org/copyleft/copyleft.html">copyleft</a>, and
|
|||
|
|
an official part of the <a href="https://www.gnu.org/">GNU</a> project (“GNU S”).</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-obtain-vegan-and-r">How to obtain <strong>vegan</strong> and R?</h3>
|
|||
|
|
<p>Both R and latest release version of <strong>vegan</strong> can be obtained through
|
|||
|
|
<a href="https://cran.r-project.org">CRAN</a>. Unstable development version of
|
|||
|
|
<strong>vegan</strong> can be obtained through
|
|||
|
|
<a href="https://github.com/vegandevs/vegan">GitHub</a>. The github page gives
|
|||
|
|
further instructions for obtaining and installing development versions
|
|||
|
|
of <strong>vegan</strong>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-r-packages-vegan-depends-on">What R packages <strong>vegan</strong> depends on?</h3>
|
|||
|
|
<p><strong>Vegan</strong> depends on the <strong>permute</strong> package which will provide advanced
|
|||
|
|
and flexible permutation routines for <strong>vegan</strong>. The <strong>permute</strong> package
|
|||
|
|
is developed together with <strong>vegan</strong> in
|
|||
|
|
<a href="https://github.com/gavinsimpson/permute">GitHub</a>.</p>
|
|||
|
|
<p>Some individual <strong>vegan</strong> functions depend on packages <strong>MASS</strong>,
|
|||
|
|
<strong>mgcv</strong>, <strong>parallel</strong>, <strong>cluster</strong> and <strong>lattice</strong>. <strong>Vegan</strong>
|
|||
|
|
dependence on <strong>tcltk</strong> is deprecated and will be removed in future
|
|||
|
|
releases. These all are base or recommended R packages that should be
|
|||
|
|
available in every R installation. <strong>Vegan</strong> declares these as
|
|||
|
|
suggested or imported packages, and you can install <strong>vegan</strong> and use
|
|||
|
|
most of its functions without these packages.</p>
|
|||
|
|
<p><strong>Vegan</strong> is accompanied with a supporting package <strong>vegan3d</strong> for
|
|||
|
|
three-dimensional and dynamic plotting. The <strong>vegan3d</strong> package needs
|
|||
|
|
<strong>tcltk</strong> and non-standard packages <strong>rgl</strong> and <strong>scatterplot3d</strong>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-other-packages-are-available-for-ecologists">What other packages are available for ecologists?</h3>
|
|||
|
|
<p>CRAN <a href="https://cran.r-project.org/web/views/">Task Views</a> include
|
|||
|
|
entries like <code>Environmetrics</code>, <code>Multivariate</code> and <code>Spatial</code> that
|
|||
|
|
describe several useful packages and functions. If you install R package
|
|||
|
|
<strong>ctv</strong>, you can inspect Task Views from your R session, and
|
|||
|
|
automatically install sets of most important packages.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-other-documentation-is-available-for-vegan">What other documentation is available for <strong>vegan</strong>?</h3>
|
|||
|
|
<p><strong>Vegan</strong> is a fully documented R package with standard help pages.
|
|||
|
|
These are the most authoritative sources of documentation (and as a
|
|||
|
|
last resource you can use the force and the read the source, as
|
|||
|
|
<strong>vegan</strong> is open source). <strong>Vegan</strong> package ships with other
|
|||
|
|
documents which can be read with <code>browseVignettes("vegan")</code>
|
|||
|
|
command. The documents included in the <strong>vegan</strong> package are</p>
|
|||
|
|
<ul>
|
|||
|
|
<li><strong>Vegan</strong> <code>NEWS</code> that can be accessed via <code>news()</code> command.</li>
|
|||
|
|
<li>This document (<code>FAQ-vegan</code>).</li>
|
|||
|
|
<li>Short introduction to basic ordination methods in <strong>vegan</strong>
|
|||
|
|
(<code>intro-vegan</code>).</li>
|
|||
|
|
<li>Introduction to diversity methods in <strong>vegan</strong>
|
|||
|
|
(<code>diversity-vegan</code>).</li>
|
|||
|
|
<li>Discussion on design decisions in <strong>vegan</strong> (<code>decision-vegan</code>).</li>
|
|||
|
|
<li>Description of variance partition procedures in function <code>varpart</code>
|
|||
|
|
(<code>partitioning</code>).</li>
|
|||
|
|
</ul>
|
|||
|
|
<p>Web documents outside the package include:</p>
|
|||
|
|
<ul>
|
|||
|
|
<li><a href="https://github.com/vegandevs/vegan">https://github.com/vegandevs/vegan</a>: development page.</li>
|
|||
|
|
<li><a href="https://vegandevs.github.io/vegan/">https://vegandevs.github.io/vegan/</a>: <strong>vegan</strong> homepage.</li>
|
|||
|
|
</ul>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="is-there-a-graphical-user-interface-gui-for-vegan">Is there a Graphical User Interface (GUI) for <strong>vegan</strong>?</h3>
|
|||
|
|
<p>Roeland Kindt has made package <strong>BiodiversityR</strong> which provides a GUI
|
|||
|
|
for <strong>vegan</strong>. The package is available at
|
|||
|
|
<a href="https://cran.r-project.org/package=BiodiversityR">CRAN</a>.
|
|||
|
|
It is not a mere GUI for <strong>vegan</strong>, but adds some new functions and
|
|||
|
|
complements <strong>vegan</strong> functions in order to provide a workbench for
|
|||
|
|
biodiversity analysis. You can install <strong>BiodiversityR</strong> using
|
|||
|
|
<code>install.packages("BiodiversityR")</code> or graphical package management menu
|
|||
|
|
in R. The GUI works on Windows, MacOS X and Linux.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-cite-vegan">How to cite <strong>vegan</strong>?</h3>
|
|||
|
|
<p>Use command <code>citation("vegan")</code> in R to see the recommended citation to
|
|||
|
|
be used in publications.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-build-vegan-from-sources">How to build <strong>vegan</strong> from sources?</h3>
|
|||
|
|
<p>In general, you do not need to build <strong>vegan</strong> from sources, but binary
|
|||
|
|
builds of release versions are available through
|
|||
|
|
<a href="https://cran.r-project.org/">CRAN</a> for Windows and MacOS X. If you use
|
|||
|
|
some other operating systems, you may have to use source packages.
|
|||
|
|
<strong>Vegan</strong> is a standard R package, and can be built like instructed in R
|
|||
|
|
documentation. <strong>Vegan</strong> contains source files in C and FORTRAN, and you
|
|||
|
|
need appropriate compilers (which may need more work in Windows and
|
|||
|
|
MacOS X).</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="are-there-binaries-for-devel-versions">Are there binaries for devel versions?</h3>
|
|||
|
|
<p>Binaries can be available from R Universe: see
|
|||
|
|
<a href="https://github.com/vegandevs/vegan">https://github.com/vegandevs/vegan</a> for instructions.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-report-a-bug-in-vegan">How to report a bug in <strong>vegan</strong>?</h3>
|
|||
|
|
<p>If you think you have found a bug in <strong>vegan</strong>, you should report it to
|
|||
|
|
<strong>vegan</strong> maintainers or developers. The preferred forum to report bugs
|
|||
|
|
is <a href="https://github.com/vegandevs/vegan/issues">GitHub</a>. The bug report
|
|||
|
|
should be so detailed that the bug can be replicated and corrected.
|
|||
|
|
Preferably, you should send an example that causes a bug. If it needs a
|
|||
|
|
data set that is not available in R, you should send a minimal data set
|
|||
|
|
as well. You also should paste the output or error message in your
|
|||
|
|
message. You also should specify which version of <strong>vegan</strong> you used.</p>
|
|||
|
|
<p>Bug reports are welcome: they are the only way to make <strong>vegan</strong>
|
|||
|
|
non-buggy.</p>
|
|||
|
|
<p>Please note that you shall not send bug reports to R mailing lists,
|
|||
|
|
since <strong>vegan</strong> is not a standard R package.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="is-it-a-bug-or-a-feature">Is it a bug or a feature?</h3>
|
|||
|
|
<p>It is not necessarily a bug if some function gives different results
|
|||
|
|
than you expect: That may be a deliberate design decision. It may be
|
|||
|
|
useful to check the documentation of the function to see what was the
|
|||
|
|
intended behaviour. It may also happen that function has an argument to
|
|||
|
|
switch the behaviour to match your expectation. For instance, function
|
|||
|
|
<code>vegdist</code> always calculates quantitative indices (when this is
|
|||
|
|
possible). If you expect it to calculate a binary index, you should use
|
|||
|
|
argument <code>binary = TRUE</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="can-i-contribute-to-vegan">Can I contribute to <strong>vegan</strong>?</h3>
|
|||
|
|
<p><strong>Vegan</strong> is dependent on user contribution. All feedback is welcome. If
|
|||
|
|
you have problems with <strong>vegan</strong>, it may be as simple as incomplete
|
|||
|
|
documentation, and we shall do our best to improve the documents.</p>
|
|||
|
|
<p>Feature requests also are welcome, but they are not necessarily
|
|||
|
|
fulfilled. A new feature will be added if it is easy to do and it looks
|
|||
|
|
useful, or if you submit code.</p>
|
|||
|
|
<p>If you can write code yourself, the best forum to contribute to vegan is
|
|||
|
|
<a href="https://github.com/vegandevs/vegan">GitHub</a>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h2 id="ordination">Ordination</h2>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="i-have-only-numeric-and-positive-data-but-vegan-still-complains">I have only numeric and positive data but <strong>vegan</strong> still complains</h3>
|
|||
|
|
<p>You are wrong! Computers are painfully pedantic, and if they find
|
|||
|
|
non-numeric or negative data entries, you really have them. Check your
|
|||
|
|
data! Most common reasons for non-numeric data are that row names were
|
|||
|
|
read as a non-numeric variable instead of being used as row names (check
|
|||
|
|
argument <code>row.names</code> in reading the data), or that the column names were
|
|||
|
|
interpreted as data (check argument <code>header = TRUE</code> in reading the
|
|||
|
|
data). Another common reason is that you had empty cells in your input
|
|||
|
|
data, and these were interpreted as missing values.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="can-i-analyse-binary-or-cover-class-data">Can I analyse binary or cover class data?</h3>
|
|||
|
|
<p>Yes. Most <strong>vegan</strong> methods can handle binary data or cover abundance
|
|||
|
|
data. Most statistical tests are based on permutation, and do not make
|
|||
|
|
distributional assumptions. There are some methods (mainly in diversity
|
|||
|
|
analysis) that need count data. These methods check that input data are
|
|||
|
|
integers, but they may be fooled by cover class data.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="why-dissimilarities-in-vegan-differ-from-other-sources">Why dissimilarities in <strong>vegan</strong> differ from other sources?</h3>
|
|||
|
|
<p>Most commonly the reason is that other software use presence–absence
|
|||
|
|
data whereas <strong>vegan</strong> used quantitative data. Usually <strong>vegan</strong> indices
|
|||
|
|
are quantitative, but you can use argument <code>binary = TRUE</code> to make them
|
|||
|
|
presence–absence. However, the index name is the same in both cases,
|
|||
|
|
although different names usually occur in literature. For instance,
|
|||
|
|
Jaccard index actually refers to the binary index, but <strong>vegan</strong> uses
|
|||
|
|
name <code>"jaccard"</code> for the quantitative index, too.</p>
|
|||
|
|
<p>Another reason may be that indices indeed are defined differently,
|
|||
|
|
because people use same names for different indices.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="why-nmds-stress-is-sometimes-0-1-and-sometimes-10">Why NMDS stress is sometimes 0.1 and sometimes 10?</h3>
|
|||
|
|
<p>Stress is a proportional measure of badness of fit. The proportions can
|
|||
|
|
be expressed either as parts of one or as percents. Function <code>isoMDS</code>
|
|||
|
|
(<strong>MASS</strong> package) uses percents, and function <code>monoMDS</code> (<strong>vegan</strong>
|
|||
|
|
package) uses proportions, and therefore the same stress is 100 times
|
|||
|
|
higher in <code>isoMDS</code>. The results of <code>goodness</code> function also depend on
|
|||
|
|
the definition of stress, and the same <code>goodness</code> is 100 times higher in
|
|||
|
|
<code>isoMDS</code> than in <code>monoMDS</code>. Both of these conventions are equally
|
|||
|
|
correct.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="i-get-zero-stress-but-no-repeated-solutions-in-metamds">I get zero stress but no repeated solutions in <code>metaMDS</code></h3>
|
|||
|
|
<p>The first (try 0) run of <code>metaMDS</code> starts from the metric scaling
|
|||
|
|
solution and is usually good, and most sofware only return that
|
|||
|
|
solution. However, <code>metaMDS</code> tries to see if that standard solution
|
|||
|
|
can be repeated, or improved and the improved solution still
|
|||
|
|
repeated. In all cases, it will return the best solution found, and
|
|||
|
|
there is no burning need to do anything if you get the message tha the
|
|||
|
|
solution could not be repeated. If you are keen to know that the
|
|||
|
|
solution really is the global optimum, you may follow the instructions
|
|||
|
|
in the <code>metaMDS</code> help section “Results Could Not Be Repeated” and try
|
|||
|
|
more.</p>
|
|||
|
|
<p>Most common reason is that you have too few observations for your NMDS.
|
|||
|
|
For <code>n</code> observations (points) and <code>k</code> dimensions you need to estimate
|
|||
|
|
<code>n*k</code> parameters (ordination scores) using <code>n*(n-1)/2</code> dissimilarities.
|
|||
|
|
For <code>k</code> dimensions you must have <code>n > 2*k + 1</code>, or for two dimensions at
|
|||
|
|
least six points. In some degenerate situations you may need even a
|
|||
|
|
larger number of points. If you have a lower number of points, you can
|
|||
|
|
find an undefined number of perfect (stress is zero) but different
|
|||
|
|
solutions. Conventional wisdom due to Kruskal is that you should have
|
|||
|
|
<code>n > 4*k + 1</code> points for <code>k</code> dimensions. A typical symptom of
|
|||
|
|
insufficient data is that you have (nearly) zero stress but no two
|
|||
|
|
convergent solutions. In those cases you should reduce the number of
|
|||
|
|
dimensions (<code>k</code>) and with very small data sets you should not use
|
|||
|
|
<code>NMDS</code>, but rely on metric methods.</p>
|
|||
|
|
<p>It seems that local and hybrid scaling with <code>monoMDS</code> have similar lower
|
|||
|
|
limits in practice (although theoretically they could differ). However,
|
|||
|
|
higher number of dimensions can be used in metric scaling, both with
|
|||
|
|
<code>monoMDS</code> and in principal coordinates analysis (<code>cmdscale</code> in
|
|||
|
|
<strong>stats</strong>, <code>wcmdscale</code> in <strong>vegan</strong>).</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="zero-dissimilarities-in-isomds">Zero dissimilarities in isoMDS</h3>
|
|||
|
|
<p>Function <code>metaMDS</code> uses function <code>monoMDS</code> as its default method for
|
|||
|
|
NMDS, and this function can handle zero dissimilarities. Alternative
|
|||
|
|
function <code>isoMDS</code> cannot handle zero dissimilarities. If you want to use
|
|||
|
|
<code>isoMDS</code>, you can use argument <code>zerodist = "add"</code> in <code>metaMDS</code> to handle
|
|||
|
|
zero dissimilarities. With this argument, zero dissimilarities are
|
|||
|
|
replaced with a small positive value, and they can be handled in
|
|||
|
|
<code>isoMDS</code>. This is a kluge, and some people do not like this. A more
|
|||
|
|
principal solution is to remove duplicate sites using R command
|
|||
|
|
<code>unique</code>. However, after some standardizations or with some
|
|||
|
|
dissimilarity indices, originally non-unique sites can have zero
|
|||
|
|
dissimilarity, and you have to resort to the kluge (or work harder with
|
|||
|
|
your data). Usually it is better to use <code>monoMDS</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="i-have-heard-that-you-cannot-fit-environmental-vectors-or-surfaces-to-nmds-results-which-only-have-rank-order-scores">I have heard that you cannot fit environmental vectors or surfaces to NMDS results which only have rank-order scores</h3>
|
|||
|
|
<p>Claims like this have indeed been at large in the Internet, but they are
|
|||
|
|
based on grave misunderstanding and are plainly wrong. NMDS ordination
|
|||
|
|
results are strictly metric, and in <strong>vegan</strong> <code>metaMDS</code> and <code>monoMDS</code>
|
|||
|
|
they are even strictly Euclidean. The method is called “non-metric”
|
|||
|
|
because the Euclidean distances in ordination space have a non-metric
|
|||
|
|
rank-order relationship to community dissimilarities. You can inspect
|
|||
|
|
this non-linear step curve using function <code>stressplot</code> in <strong>vegan</strong>.
|
|||
|
|
Because the ordination scores are strictly Euclidean, it is correct to
|
|||
|
|
use <strong>vegan</strong> functions <code>envfit</code> and <code>ordisurf</code> with NMDS results.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="where-can-i-find-numerical-scores-of-ordination-axes">Where can I find numerical scores of ordination axes?</h3>
|
|||
|
|
<p>Normally you can use function <code>scores</code> to extract ordination scores for
|
|||
|
|
any ordination method. The <code>scores</code> function can also find ordination
|
|||
|
|
scores for many non-<strong>vegan</strong> functions such as for <code>prcomp</code> and
|
|||
|
|
<code>princomp</code> and for some <strong>ade4</strong> functions.</p>
|
|||
|
|
<p>In some cases the ordination result object stores raw scores, and the
|
|||
|
|
axes are also scaled appropriate when you access them with <code>scores</code>. For
|
|||
|
|
instance, in <code>cca</code> and <code>rda</code> the ordination object has only so-called
|
|||
|
|
normalized scores, and they are scaled for ordination plots or for other
|
|||
|
|
use when they are accessed with <code>scores</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-the-rda-results-are-scaled">How the RDA results are scaled?</h3>
|
|||
|
|
<p>The scaling or RDA results indeed differ from most other software
|
|||
|
|
packages. The scaling of RDA is such a complicated issue that it cannot
|
|||
|
|
be explained in this FAQ, but it is explained in a separate pdf document
|
|||
|
|
on “Design decision and implementation details in vegan” that you can
|
|||
|
|
read with command <code>browseVignettes("vegan")</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="i-cannot-print-and-plot-rda-results-properly">I cannot print and plot RDA results properly</h3>
|
|||
|
|
<p>If the RDA ordination results have a weird format or you cannot plot
|
|||
|
|
them properly, you probably have a name clash with <strong>klaR</strong> package
|
|||
|
|
which also has function <code>rda</code>, and the <strong>klaR</strong> <code>print</code>, <code>plot</code> or
|
|||
|
|
<code>predict</code> functions are used for <strong>vegan</strong> RDA results. You can choose
|
|||
|
|
between <code>rda</code> functions using <code>vegan::rda()</code> or <code>klaR::rda()</code>: you
|
|||
|
|
will get obscure error messages if you use the wrong function. In
|
|||
|
|
general, <strong>vegan</strong> should be able to work normally if <strong>vegan</strong> was
|
|||
|
|
loaded after <strong>klaR</strong>, but if <strong>klaR</strong> was loaded later, its functions
|
|||
|
|
will take precedence over <strong>vegan</strong>. Sometimes <strong>vegan</strong> namespace is
|
|||
|
|
loaded automatically when restoring a previously stored workspace at
|
|||
|
|
the start-up, and then <strong>klaR</strong> methods will always take precedence
|
|||
|
|
over <strong>vegan</strong>. You should check your loaded packages. <strong>klaR</strong> may be
|
|||
|
|
also loaded indirectly via other packages (in the reported cases it
|
|||
|
|
was most often loaded via <strong>agricolae</strong> package). <strong>Vegan</strong> and
|
|||
|
|
<strong>klaR</strong> both have the same function name (<code>rda</code>), and it may not be
|
|||
|
|
possible to use these packages simultaneously, and the safest choice
|
|||
|
|
is to unload one of the packages if only possible. See discussion in
|
|||
|
|
<a href="https://github.com/vegandevs/vegan/issues/277">vegan github issues</a>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="ordination-fails-with-error-in-la-svd">Ordination fails with “Error in La.svd”</h3>
|
|||
|
|
<p>Constrained ordination (<code>cca</code>, <code>rda</code>, <code>dbrda</code>, <code>capscale</code>) will
|
|||
|
|
sometimes fail with error message
|
|||
|
|
<code>Error in La.svd(x, nu, nv): error code 1 from Lapack routine 'dgesdd'.</code></p>
|
|||
|
|
<p>It seems that the basic problem is in the <code>svd</code> function of <code>LAPACK</code>
|
|||
|
|
that is used for numerical analysis in R. <code>LAPACK</code> is an external
|
|||
|
|
library that is beyond the control of package developers and R core team
|
|||
|
|
so that these problems may be unsolvable.</p>
|
|||
|
|
<p>Reducing the range of constraints (environmental variables) helps
|
|||
|
|
sometimes. For instance, multiplying constraints by a constant < 1.
|
|||
|
|
This rescaling does not influence the numerical results of constrained
|
|||
|
|
ordination, but it can complicate further analyses when values of
|
|||
|
|
constraints are needed, because the same scaling must be applied
|
|||
|
|
there. The reports on the problems are getting rare and it may that
|
|||
|
|
this problem is fixed in R and <code>LAPACK</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="variance-explained-by-ordination-axes">Variance explained by ordination axes.</h3>
|
|||
|
|
<p>In general, <strong>vegan</strong> does not directly give any statistics on the
|
|||
|
|
“variance explained” by ordination axes or by the constrained axes. This
|
|||
|
|
is a design decision: I think this information is normally useless and
|
|||
|
|
often misleading. In community ordination, the goal typically is not to
|
|||
|
|
explain the variance, but to find the “gradients” or main trends in the
|
|||
|
|
data. The “total variation” often is meaningless, and all proportions of
|
|||
|
|
meaningless values also are meaningless. Often a better solution
|
|||
|
|
explains a smaller part of “total variation”. For instance, in
|
|||
|
|
unstandardized principal components analysis most of the variance is
|
|||
|
|
generated by a small number of most abundant species, and they are easy
|
|||
|
|
to “explain” because data really are not very multivariate. If you
|
|||
|
|
standardize your data, all species are equally important. The first axes
|
|||
|
|
explains much less of the “total variation”, but now they explain all
|
|||
|
|
species equally, and results typically are much more useful for the
|
|||
|
|
whole community. Correspondence analysis uses another measure of
|
|||
|
|
variation (which is not variance), and again it typically explains a
|
|||
|
|
“smaller proportion” than principal components but with a better result.
|
|||
|
|
Detrended correspondence analysis and nonmetric multidimensional scaling
|
|||
|
|
even do not try to “explain” the variation, but use other criteria. All
|
|||
|
|
methods are incommensurable, and it is impossible to compare methods
|
|||
|
|
using “explanation of variation”.</p>
|
|||
|
|
<p>If you still want to get “explanation of variation” (or a deranged
|
|||
|
|
editor requests that from you), it is possible to get this information
|
|||
|
|
for some methods:</p>
|
|||
|
|
<ul>
|
|||
|
|
<li>
|
|||
|
|
<p>Eigenvector methods: Functions <code>rda</code>, <code>cca</code>, <code>dbrda</code> and <code>capscale</code>
|
|||
|
|
give the variation of conditional (partialled), constrained
|
|||
|
|
(canonical) and residual components. Function <code>eigenvals</code> extracts
|
|||
|
|
the eigenvalues, and <code>summary(eigenvals(ord))</code> reports the
|
|||
|
|
proportions explained in the result object <code>ord</code>, and also works
|
|||
|
|
with <code>decorana</code> and <code>wcmdscale</code>. Function <code>RsquareAdj</code> gives the
|
|||
|
|
R-squared and adjusted R-squared (if available) for constrained
|
|||
|
|
components. Function <code>goodness</code> gives the same statistics for
|
|||
|
|
individual species or sites. In addition, there is a special
|
|||
|
|
function <code>varpart</code> for unbiased partitioning of variance between up
|
|||
|
|
to four separate components in redundancy analysis.</p>
|
|||
|
|
</li>
|
|||
|
|
<li>
|
|||
|
|
<p>Nonmetric multidimensional scaling. NMDS is a method for
|
|||
|
|
nonlinear mapping, and the concept of of variation explained does
|
|||
|
|
not make sense. However, 1 - stress^2 transforms nonlinear stress
|
|||
|
|
into quantity analogous to squared correlation coefficient. Function
|
|||
|
|
<code>stressplot</code> displays the nonlinear fit and gives this statistic.</p>
|
|||
|
|
</li>
|
|||
|
|
</ul>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="can-i-have-random-effects-in-constrained-ordination-or-in-adonis">Can I have random effects in constrained ordination or in <code>adonis</code>?</h3>
|
|||
|
|
<p>No. Strictly speaking, this is impossible. However, you can define
|
|||
|
|
models that respond to similar goals as random effects models, although
|
|||
|
|
they strictly speaking use only fixed effects.</p>
|
|||
|
|
<p>Constrained ordination functions <code>cca</code>, <code>rda</code> and <code>dbrda</code> can have
|
|||
|
|
<code>Condition()</code> terms in their formula. The <code>Condition()</code> define partial
|
|||
|
|
terms that are fitted before other constraints and can be used to remove
|
|||
|
|
the effects of background variables, and their contribution to
|
|||
|
|
decomposing inertia (variance) is reported separately. These partial
|
|||
|
|
terms are often regarded as similar to random effects, but they are
|
|||
|
|
still fitted in the same way as other terms and strictly speaking they
|
|||
|
|
are fixed terms.</p>
|
|||
|
|
<p>Function <code>adonis2</code> can evaluate terms sequentially. In a model with
|
|||
|
|
right-hand-side <code>~ A + B</code> the effects of <code>A</code> are evaluated first, and
|
|||
|
|
the effects of <code>B</code> after removing the effects of <code>A</code>. Sequential tests
|
|||
|
|
are also available in <code>anova</code> function for constrained ordination
|
|||
|
|
results by setting argument <code>by = "term"</code>. In this way, the first terms
|
|||
|
|
can serve in a similar role as random effects, although they are fitted
|
|||
|
|
in the same way as all other terms, and strictly speaking they are fixed
|
|||
|
|
terms.</p>
|
|||
|
|
<p>All permutation tests in <strong>vegan</strong> are based on the <strong>permute</strong> package
|
|||
|
|
that allows constructing various restricted permutation schemes. For
|
|||
|
|
instance, you can set levels of <code>plots</code> or <code>blocks</code> for a factor
|
|||
|
|
regarded as a random term.</p>
|
|||
|
|
<p>A major reason why real random effects models are impossible in most
|
|||
|
|
<strong>vegan</strong> functions is that their tests are based on the permutation of
|
|||
|
|
the data. The data are given, that is fixed, and therefore permutation
|
|||
|
|
tests are basically tests of fixed terms on fixed data. Random effect
|
|||
|
|
terms would require permutations of data with a random component instead
|
|||
|
|
of the given, fixed data, and such tests are not available in <strong>vegan</strong>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="is-it-possible-to-have-passive-points-in-ordination">Is it possible to have passive points in ordination?</h3>
|
|||
|
|
<p><strong>Vegan</strong> does not have a concept of passive points, or a point that
|
|||
|
|
should only little influence the ordination results. However, you can
|
|||
|
|
add points to eigenvector methods using <code>predict</code> functions with
|
|||
|
|
<code>newdata</code>. You can first perform an ordination without some species or
|
|||
|
|
sites, and then you can find scores for all points using your complete
|
|||
|
|
data as <code>newdata</code>. The <code>predict</code> functions are available for basic
|
|||
|
|
eigenvector methods in <strong>vegan</strong> (<code>cca</code>, <code>rda</code>, <code>decorana</code>, for an
|
|||
|
|
up-to-date list, use command <code>methods("predict")</code>).</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="class-variables-and-dummies">Class variables and dummies</h3>
|
|||
|
|
<p>You should define a class variable as an R <code>factor</code>, and <strong>vegan</strong> will
|
|||
|
|
automatically handle them.</p>
|
|||
|
|
<p>R (and <strong>vegan</strong>) knows both unordered and ordered factors. Unordered
|
|||
|
|
factors are internally coded as dummy variables, but one redundant level
|
|||
|
|
is removed or aliased. With default contrasts, the removed level is the
|
|||
|
|
first one. Ordered factors are expressed as polynomial contrasts. Both
|
|||
|
|
of these contrasts explained in standard R documentation.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-are-environmental-arrows-scaled">How are environmental arrows scaled?</h3>
|
|||
|
|
<p>The printed output of <code>envfit</code> gives the direction cosines which are the
|
|||
|
|
coordinates of unit length arrows. For plotting, these are scaled by
|
|||
|
|
their correlation (square roots of column <code>r2</code>). You can see the scaled
|
|||
|
|
lengths of <code>envfit</code> arrows using command <code>scores</code>.</p>
|
|||
|
|
<p>The scaled environmental vectors from <code>envfit</code> and the arrows for
|
|||
|
|
continuous environmental variables in constrained ordination (<code>cca</code>,
|
|||
|
|
<code>rda</code>, <code>dbrda</code>) are adjusted to fill the current graph. The lengths
|
|||
|
|
of arrows do not have fixed meaning with respect to the points (species,
|
|||
|
|
sites), but they can only compared against each other, and therefore
|
|||
|
|
only their relative lengths are important.</p>
|
|||
|
|
<p>If you want change the scaling of the arrows, you can use <code>text</code>
|
|||
|
|
(plotting arrows and text) or <code>points</code> (plotting only arrows) functions
|
|||
|
|
for constrained ordination. These functions have argument <code>arrow.mul</code>
|
|||
|
|
which sets the multiplier. The <code>plot</code> function for <code>envfit</code> also has the
|
|||
|
|
<code>arrow.mul</code> argument to set the arrow multiplier. If you save the
|
|||
|
|
invisible result of the constrained ordination <code>plot</code> command, you can
|
|||
|
|
see the value of the currently used <code>arrow.mul</code> which is saved as an
|
|||
|
|
attribute of <code>biplot</code> scores.</p>
|
|||
|
|
<p>Function <code>ordiArrowMul</code> is used to find the scaling for the current
|
|||
|
|
plot. You can use this function to see how arrows would be scaled:</p>
|
|||
|
|
<pre><code class="language-r">sol <- cca(varespec)
|
|||
|
|
ef <- envfit(sol ~ ., varechem)
|
|||
|
|
plot(sol)
|
|||
|
|
ordiArrowMul(scores(ef, display="vectors"))
|
|||
|
|
</code></pre>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="i-want-to-use-helmert-or-sum-contrasts">I want to use Helmert or sum contrasts</h3>
|
|||
|
|
<p><code>vegan</code> uses standard R utilities for defining contrasts. The default in
|
|||
|
|
standard installations is to use treatment contrasts, but you can change
|
|||
|
|
the behaviour globally setting <code>options</code> or locally by using keyword
|
|||
|
|
<code>contrasts</code>. Please check the R help pages and user manuals for details.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="what-are-aliased-variables-and-how-to-see-them">What are aliased variables and how to see them?</h3>
|
|||
|
|
<p>Aliased variable has no information because it can be expressed with the
|
|||
|
|
help of other variables. Such variables are automatically removed in
|
|||
|
|
constrained ordination in <strong>vegan</strong>. The aliased variables can be
|
|||
|
|
redundant levels of factors or whole variables.</p>
|
|||
|
|
<p><strong>Vegan</strong> function <code>alias</code> gives the defining equations for aliased
|
|||
|
|
variables. If you only want to see the names of aliased variables or
|
|||
|
|
levels in solution <code>sol</code>, use <code>alias(sol, names.only=TRUE)</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="plotting-aliased-variables">Plotting aliased variables</h3>
|
|||
|
|
<p>You can fit vectors or class centroids for aliased variables using
|
|||
|
|
<code>envfit</code> function. The <code>envfit</code> function uses weighted fitting, and the
|
|||
|
|
fitted vectors are identical to the vectors in correspondence analysis.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="restricted-permutations-in-vegan">Restricted permutations in <strong>vegan</strong></h3>
|
|||
|
|
<p><strong>Vegan</strong> uses <strong>permute</strong> package in all its permutation tests. The
|
|||
|
|
<strong>permute</strong> package will allow restricted permutation designs for time
|
|||
|
|
series, line transects, spatial grids and blocking factors. The
|
|||
|
|
construction of restricted permutation schemes is explained in the
|
|||
|
|
manual page <code>permutations</code> in <strong>vegan</strong> and in the documentation of the
|
|||
|
|
<strong>permute</strong> package.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-use-different-plotting-symbols-in-ordination-graphics">How to use different plotting symbols in ordination graphics?</h3>
|
|||
|
|
<p>The default ordination <code>plot</code> function is intended for fast plotting and
|
|||
|
|
it is not very configurable. To use different plotting symbols, you
|
|||
|
|
should first create and empty ordination plot with
|
|||
|
|
<code>plot(..., type="n")</code>, and then add <code>points</code> or <code>text</code> to the created
|
|||
|
|
empty frame (here <code>...</code> means other arguments you want to give to your
|
|||
|
|
<code>plot</code> command). The <code>points</code> and <code>text</code> commands are fully
|
|||
|
|
configurable, and allow different plotting symbols and characters.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-to-avoid-cluttered-ordination-graphs">How to avoid cluttered ordination graphs?</h3>
|
|||
|
|
<p>If there is a really high number of species or sites, the graphs often
|
|||
|
|
are congested and many labels are overwritten. It may be impossible to
|
|||
|
|
have complete readable graphics with some data sets. Below we give a
|
|||
|
|
brief overview of tricks you can use. Gavin Simpson’s blog <a href="https://fromthebottomoftheheap.net">From the
|
|||
|
|
bottom of the heap</a> has a series
|
|||
|
|
of articles on “decluttering ordination plots” with more detailed
|
|||
|
|
discussion and examples.</p>
|
|||
|
|
<ul>
|
|||
|
|
<li>Use only points, possibly with different types if you do not need to
|
|||
|
|
see the labels. You may need to first create an empty plot using
|
|||
|
|
<code>plot(..., type="n")</code>, if you are not satisfied with the default
|
|||
|
|
graph. (Here and below <code>...</code> means other arguments you want to give
|
|||
|
|
to your <code>plot</code> command.)</li>
|
|||
|
|
<li>Use points and add labels to desired points using interactive
|
|||
|
|
<code>identify</code> command if you do not need to see all labels.</li>
|
|||
|
|
<li>Add labels using function <code>ordilabel</code> which uses non-transparent
|
|||
|
|
background to the text. The labels still shadow each other, but the
|
|||
|
|
uppermost labels are readable. Argument <code>priority</code> will help in
|
|||
|
|
displaying the most interesting labels (see <a href="https://fromthebottomoftheheap.net/2013/01/12/decluttering-ordination-plots-in-vegan-part-1-ordilabel/">Decluttering blog, part
|
|||
|
|
1</a>).</li>
|
|||
|
|
<li>Use <code>orditorp</code> function that uses labels only if these can be added
|
|||
|
|
to a graph without overwriting other labels, and points otherwise,
|
|||
|
|
if you do not need to see all labels. You must first create an empty
|
|||
|
|
plot using <code>plot(..., type="n")</code>, and then add labels or points with
|
|||
|
|
<code>orditorp</code> (see <a href="https://fromthebottomoftheheap.net/2013/01/13/decluttering-ordination-plots-in-vegan-part-2-orditorp/">Decluttering
|
|||
|
|
blog</a>).</li>
|
|||
|
|
<li>Use <code>ordipointlabel</code> which uses points and text labels to the
|
|||
|
|
points, and tries to optimize the location of the text to minimize
|
|||
|
|
the overlap (see <a href="https://fromthebottomoftheheap.net/2013/06/27/decluttering-ordination-plots-in-vegan-part-3-ordipointlabel/">Decluttering
|
|||
|
|
blog</a>).</li>
|
|||
|
|
<li>Ordination <code>text</code> and <code>points</code> functions have argument <code>select</code> that
|
|||
|
|
can be used for full control of selecting items plotted as text or
|
|||
|
|
points.</li>
|
|||
|
|
<li>Use interactive <code>orditkplot</code> function (<strong>vegan3d</strong> package)
|
|||
|
|
that lets you drag labels of points to better positions if you need to see
|
|||
|
|
all labels. Only one set of points can be used
|
|||
|
|
(see <a href="https://fromthebottomoftheheap.net/2013/12/31/decluttering-ordination-in-vegan-part-4-orditkplot/">Decluttering blog</a>).</li>
|
|||
|
|
<li>Most <code>plot</code> functions allow you to zoom to a part of the graph using
|
|||
|
|
<code>xlim</code> and <code>ylim</code> arguments to reduce clutter in congested areas.</li>
|
|||
|
|
</ul>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="can-i-flip-an-axis-in-ordination-diagram">Can I flip an axis in ordination diagram?</h3>
|
|||
|
|
<p>Use <code>xlim</code> or <code>ylim</code> with flipped limits. If you have model
|
|||
|
|
<code>mod <- cca(dune)</code> you can flip the first axis with
|
|||
|
|
<code>plot(mod, xlim = c(3, -2))</code>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="can-i-zoom-into-an-ordination-plot">Can I zoom into an ordination plot?</h3>
|
|||
|
|
<p>You can use <code>xlim</code> and <code>ylim</code> arguments in <code>plot</code> or <code>ordiplot</code> to zoom
|
|||
|
|
into ordination diagrams. Normally you must set both <code>xlim</code> and <code>ylim</code>
|
|||
|
|
because ordination plots will keep the equal aspect ratio of axes, and
|
|||
|
|
they will fill the graph so that the longer axis will fit.</p>
|
|||
|
|
<p>Dynamic zooming can be done with function <code>orditkplot</code> in CRAN package
|
|||
|
|
<strong>vegan3d</strong>. You can directly save the edited <code>orditkplot</code> graph in
|
|||
|
|
various graphic formats, or you can export the graph object back to R
|
|||
|
|
session and use <code>plot</code> to display the results.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h2 id="other-analysis-methods">Other analysis methods</h2>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="is-there-twinspan">Is there TWINSPAN?</h3>
|
|||
|
|
<p>TWINSPAN for R is available in
|
|||
|
|
<a href="https://github.com/jarioksa/twinspan">github</a>.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="why-restricted-permutation-does-not-influence-adonis-results">Why restricted permutation does not influence adonis results?</h3>
|
|||
|
|
<p>The permutation scheme influences the permutation distribution of the
|
|||
|
|
statistics and probably the significance levels, but does not influence
|
|||
|
|
the calculation of the statistics.</p>
|
|||
|
|
<hr />
|
|||
|
|
<h3 id="how-is-deviance-calculated">How is deviance calculated?</h3>
|
|||
|
|
<p>Some <strong>vegan</strong> functions, such as <code>radfit</code> use base R facility of
|
|||
|
|
<code>family</code> in maximum likelihood estimation. This allows use of several
|
|||
|
|
alternative error distributions, among them <code>"poisson"</code> and
|
|||
|
|
<code>"gaussian"</code>. The R <code>family</code> also defines the deviance. You can see the
|
|||
|
|
equations for deviance with commands like <code>poisson()$dev</code> or
|
|||
|
|
<code>gaussian()$dev</code>.</p>
|
|||
|
|
<p>In general, deviance is 2 times log.likelihood shifted so that models
|
|||
|
|
with exact fit have zero deviance.</p>
|
|||
|
|
<hr />
|
|||
|
|
</div>
|
|||
|
|
<script src="https://cdn.jsdelivr.net/npm/prismjs@1.29.0/components/prism-core.min.js" defer></script>
|
|||
|
|
<script src="https://cdn.jsdelivr.net/npm/prismjs@1.29.0/plugins/autoloader/prism-autoloader.min.js" defer></script>
|
|||
|
|
</body>
|
|||
|
|
</html>
|