<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<title>add_err. mkin 0.9.44</title>
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<meta name="author" content="
Johannes Ranke
">
<link href="css/bootstrap.css" rel="stylesheet">
<link href="css/bootstrap-responsive.css" rel="stylesheet">
<link href="css/highlight.css" rel="stylesheet">
<link href="css/staticdocs.css" rel="stylesheet">
<!--[if lt IE 9]>
<script src="http://html5shim.googlecode.com/svn/trunk/html5.js"></script>
<![endif]-->
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
tex2jax: {
inlineMath: [ ['$','$'], ["\\(","\\)"] ],
processEscapes: true
}
});
</script>
<script type="text/javascript"
src="http://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML">
</script>
</head>
<body>
<div class="navbar">
<div class="navbar-inner">
<div class="container">
<a class="brand" href="#">mkin 0.9.44</a>
<div class="nav">
<ul class="nav">
<li><a href="index.html"><i class="icon-home icon-white"></i> Index</a></li>
</ul>
</div>
</div>
</div>
</div>
<div class="container">
<header>
</header>
<h1>
Add normally distributed errors to simulated kinetic degradation data
</h1>
<div class="row">
<div class="span8">
<h2>Usage</h2>
<pre><div>add_err(prediction, sdfunc, n = 1000, LOD = 0.1, reps = 2, digits = 1, seed = NA)</div></pre>
<h2>Arguments</h2>
<dl>
<dt>prediction</dt>
<dd>
A prediction from a kinetic model as produced by <code><a href='mkinpredict.html'>mkinpredict</a></code>.
</dd>
<dt>sdfunc</dt>
<dd>
A function taking the predicted value as its only argument and returning
a standard deviation that should be used for generating the random error
terms for this value.
</dd>
<dt>n</dt>
<dd>
The number of datasets to be generated.
</dd>
<dt>LOD</dt>
<dd>
The limit of detection (LOD). Values that are below the LOD after adding
the random error will be set to NA.
</dd>
<dt>reps</dt>
<dd>
The number of replicates to be generated within the datasets.
</dd>
<dt>digits</dt>
<dd>
The number of digits to which the values will be rounded.
</dd>
<dt>seed</dt>
<dd>
The seed used for the generation of random numbers. If NA, the seed
is not set.
</dd>
</dl>
<div class="Description">
<h2>Description</h2>
<p>Normally distributed errors are added to data predicted for a specific
degradation model using <code><a href='mkinpredict.html'>mkinpredict</a></code>. The variance of the error
may depend on the predicted value and is specified as a standard deviation.</p>
</div>
<div class="Value">
<h2>Value</h2>
<p><dl>
A list of datasets compatible with <code><a href='mmkin.html'>mmkin</a></code>, i.e.
the components of the list are datasets compatible with
<code><a href='mkinfit.html'>mkinfit</a></code>.
</dl></p>
</div>
<div class="References">
<h2>References</h2>
<p>Ranke J and Lehmann R (2015) To t-test or not to t-test, that is the question. XV Symposium on Pesticide Chemistry 2-4 September 2015, Piacenza, Italy
http://chem.uft.uni-bremen.de/ranke/posters/piacenza_2015.pdf</p>
</div>
<h2 id="examples">Examples</h2>
<pre class="examples"><div class='input'># The kinetic model
m_SFO_SFO <- mkinmod(parent = mkinsub("SFO", "M1"),
M1 = mkinsub("SFO"), use_of_ff = "max")
</div>
<strong class='message'>Successfully compiled differential equation model from auto-generated C code.</strong>
<div class='input'>
# Generate a prediction for a specific set of parameters
sampling_times = c(0, 1, 3, 7, 14, 28, 60, 90, 120)
# This is the prediction used for the "Type 2 datasets" on the Piacenza poster
# from 2015
d_SFO_SFO <- mkinpredict(m_SFO_SFO,
c(k_parent = 0.1, f_parent_to_M1 = 0.5,
k_M1 = log(2)/1000),
c(parent = 100, M1 = 0),
sampling_times)
# Add an error term with a constant (independent of the value) standard deviation
# of 10, and generate three datasets
d_SFO_SFO_err <- add_err(d_SFO_SFO, function(x) 10, n = 3, seed = 123456789 )
# Name the datasets for nicer plotting
names(d_SFO_SFO_err) <- paste("Dataset", 1:3)
# Name the model in the list of models (with only one member in this case)
# for nicer plotting later on.
# Be quiet and use the faster Levenberg-Marquardt algorithm, as the datasets
# are easy and examples are run often. Use only one core not to offend CRAN
# checks
f_SFO_SFO <- mmkin(list("SFO-SFO" = m_SFO_SFO),
d_SFO_SFO_err, cores = 1,
quiet = TRUE, method.modFit = "Marq")
plot(f_SFO_SFO)
</div>
<p><img src='add_err-4.png' alt='' width='540' height='400' /></p>
<div class='input'>
# We would like to inspect the fit for dataset 3 more closely
# Using double brackets makes the returned object an mkinfit object
# instead of a list of mkinfit objects, so plot.mkinfit is used
plot(f_SFO_SFO[[3]], show_residuals = TRUE)
</div>
<p><img src='add_err-6.png' alt='' width='540' height='400' /></p>
<div class='input'>
# If we use single brackets, we should give two indices (model and dataset),
# and plot.mmkin is used
plot(f_SFO_SFO[1, 3])
</div>
<p><img src='add_err-8.png' alt='' width='540' height='400' /></p>
<div class='input'></div></pre>
</div>
<div class="span4">
<!-- <ul>
<li>add_err</li>
</ul>
<ul>
<li> manip </li>
</ul> -->
<h2>Author</h2>
Johannes Ranke
</div>
</div>
<footer>
<p class="pull-right"><a href="#">Back to top</a></p>
<p>Built by <a href="https://github.com/hadley/staticdocs">staticdocs</a>. Styled with <a href="http://twitter.github.com/bootstrap">bootstrap</a>.</p>
</footer>
</div>
</body>
</html>
