<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta charset="utf-8" />
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="pandoc" />
<meta name="author" content="Yuxin Zou" />
<meta name="date" content="2018-07-25" />
<title>Estimate cor—max MASH</title>
<script src="site_libs/jquery-1.11.3/jquery.min.js"></script>
<meta name="viewport" content="width=device-width, initial-scale=1" />
<link href="site_libs/bootstrap-3.3.5/css/cosmo.min.css" rel="stylesheet" />
<script src="site_libs/bootstrap-3.3.5/js/bootstrap.min.js"></script>
<script src="site_libs/bootstrap-3.3.5/shim/html5shiv.min.js"></script>
<script src="site_libs/bootstrap-3.3.5/shim/respond.min.js"></script>
<script src="site_libs/jqueryui-1.11.4/jquery-ui.min.js"></script>
<link href="site_libs/tocify-1.9.1/jquery.tocify.css" rel="stylesheet" />
<script src="site_libs/tocify-1.9.1/jquery.tocify.js"></script>
<script src="site_libs/navigation-1.1/tabsets.js"></script>
<script src="site_libs/navigation-1.1/codefolding.js"></script>
<link href="site_libs/highlightjs-9.12.0/textmate.css" rel="stylesheet" />
<script src="site_libs/highlightjs-9.12.0/highlight.js"></script>
<script src="site_libs/kePrint-0.0.1/kePrint.js"></script>
<link href="site_libs/font-awesome-5.0.13/css/fa-svg-with-js.css" rel="stylesheet" />
<script src="site_libs/font-awesome-5.0.13/js/fontawesome-all.min.js"></script>
<script src="site_libs/font-awesome-5.0.13/js/fa-v4-shims.min.js"></script>
<style type="text/css">code{white-space: pre;}</style>
<style type="text/css">
pre:not([class]) {
background-color: white;
}
</style>
<script type="text/javascript">
if (window.hljs) {
hljs.configure({languages: []});
hljs.initHighlightingOnLoad();
if (document.readyState && document.readyState === "complete") {
window.setTimeout(function() { hljs.initHighlighting(); }, 0);
}
}
</script>
<style type="text/css">
h1 {
font-size: 34px;
}
h1.title {
font-size: 38px;
}
h2 {
font-size: 30px;
}
h3 {
font-size: 24px;
}
h4 {
font-size: 18px;
}
h5 {
font-size: 16px;
}
h6 {
font-size: 12px;
}
.table th:not([align]) {
text-align: left;
}
</style>
</head>
<body>
<style type = "text/css">
.main-container {
max-width: 940px;
margin-left: auto;
margin-right: auto;
}
code {
color: inherit;
background-color: rgba(0, 0, 0, 0.04);
}
img {
max-width:100%;
height: auto;
}
.tabbed-pane {
padding-top: 12px;
}
.html-widget {
margin-bottom: 20px;
}
button.code-folding-btn:focus {
outline: none;
}
</style>
<style type="text/css">
/* padding for bootstrap navbar */
body {
padding-top: 51px;
padding-bottom: 40px;
}
/* offset scroll position for anchor links (for fixed navbar) */
.section h1 {
padding-top: 56px;
margin-top: -56px;
}
.section h2 {
padding-top: 56px;
margin-top: -56px;
}
.section h3 {
padding-top: 56px;
margin-top: -56px;
}
.section h4 {
padding-top: 56px;
margin-top: -56px;
}
.section h5 {
padding-top: 56px;
margin-top: -56px;
}
.section h6 {
padding-top: 56px;
margin-top: -56px;
}
</style>
<script>
// manage active state of menu based on current page
$(document).ready(function () {
// active menu anchor
href = window.location.pathname
href = href.substr(href.lastIndexOf('/') + 1)
if (href === "")
href = "index.html";
var menuAnchor = $('a[href="' + href + '"]');
// mark it active
menuAnchor.parent().addClass('active');
// if it's got a parent navbar menu mark it active as well
menuAnchor.closest('li.dropdown').addClass('active');
});
</script>
<div class="container-fluid main-container">
<!-- tabsets -->
<script>
$(document).ready(function () {
window.buildTabsets("TOC");
});
</script>
<!-- code folding -->
<style type="text/css">
.code-folding-btn { margin-bottom: 4px; }
</style>
<script>
$(document).ready(function () {
window.initializeCodeFolding("hide" === "show");
});
</script>
<script>
$(document).ready(function () {
// move toc-ignore selectors from section div to header
$('div.section.toc-ignore')
.removeClass('toc-ignore')
.children('h1,h2,h3,h4,h5').addClass('toc-ignore');
// establish options
var options = {
selectors: "h1,h2,h3",
theme: "bootstrap3",
context: '.toc-content',
hashGenerator: function (text) {
return text.replace(/[.\\/?&!#<>]/g, '').replace(/\s/g, '_').toLowerCase();
},
ignoreSelector: ".toc-ignore",
scrollTo: 0
};
options.showAndHide = true;
options.smoothScroll = true;
// tocify
var toc = $("#TOC").tocify(options).data("toc-tocify");
});
</script>
<style type="text/css">
#TOC {
margin: 25px 0px 20px 0px;
}
@media (max-width: 768px) {
#TOC {
position: relative;
width: 100%;
}
}
.toc-content {
padding-left: 30px;
padding-right: 40px;
}
div.main-container {
max-width: 1200px;
}
div.tocify {
width: 20%;
max-width: 260px;
max-height: 85%;
}
@media (min-width: 768px) and (max-width: 991px) {
div.tocify {
width: 25%;
}
}
@media (max-width: 767px) {
div.tocify {
width: 100%;
max-width: none;
}
}
.tocify ul, .tocify li {
line-height: 20px;
}
.tocify-subheader .tocify-item {
font-size: 0.90em;
padding-left: 25px;
text-indent: 0;
}
.tocify .list-group-item {
border-radius: 0px;
}
</style>
<!-- setup 3col/9col grid for toc_float and main content -->
<div class="row-fluid">
<div class="col-xs-12 col-sm-4 col-md-3">
<div id="TOC" class="tocify">
</div>
</div>
<div class="toc-content col-xs-12 col-sm-8 col-md-9">
<div class="navbar navbar-default navbar-fixed-top" role="navigation">
<div class="container">
<div class="navbar-header">
<button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar">
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a class="navbar-brand" href="index.html">mash_application</a>
</div>
<div id="navbar" class="navbar-collapse collapse">
<ul class="nav navbar-nav">
<li>
<a href="index.html">Home</a>
</li>
<li>
<a href="about.html">About</a>
</li>
<li>
<a href="license.html">License</a>
</li>
</ul>
<ul class="nav navbar-nav navbar-right">
<li>
<a href="https://github.com/jdblischak/workflowr">
<span class="fa fa-github"></span>
</a>
</li>
</ul>
</div><!--/.nav-collapse -->
</div><!--/.container -->
</div><!--/.navbar -->
<!-- Add a small amount of space between sections. -->
<style type="text/css">
div.section {
padding-top: 12px;
}
</style>
<div class="fluid-row" id="header">
<div class="btn-group pull-right">
<button type="button" class="btn btn-default btn-xs dropdown-toggle" data-toggle="dropdown" aria-haspopup="true" aria-expanded="false"><span>Code</span> <span class="caret"></span></button>
<ul class="dropdown-menu" style="min-width: 50px;">
<li><a id="rmd-show-all-code" href="#">Show All Code</a></li>
<li><a id="rmd-hide-all-code" href="#">Hide All Code</a></li>
</ul>
</div>
<h1 class="title toc-ignore">Estimate cor—max MASH</h1>
<h4 class="author"><em>Yuxin Zou</em></h4>
<h4 class="date"><em>2018-07-25</em></h4>
</div>
<!-- Update knitr chunk options -->
<!-- Insert the date the file was last updated -->
<p><strong>Last updated:</strong> 2018-08-15</p>
<pre class="r"><code>library(mashr)</code></pre>
<pre><code>Loading required package: ashr</code></pre>
<pre class="r"><code>source('../code/generateDataV.R')
source('../code/estimate_cor.R')
source('../code/summary.R')
library(knitr)
library(kableExtra)</code></pre>
<p>Apply the max methods for correlation matrix on mash data. The estimated V from MLE and EM perform better than the truncated correlation (error, mash log likelihood, ROC).</p>
<div id="one-example" class="section level1">
<h1>One example</h1>
<p><span class="math display">\[
\hat{\beta}|\beta \sim N_{3}(\hat{\beta}; \beta, \left(\begin{matrix} 1 & 0.7 & 0.2 \\
0.7 & 1 & 0.4 \\
0.2 & 0.4 & 1 \end{matrix}\right))
\]</span></p>
<p><span class="math display">\[
\beta \sim \frac{1}{4}\delta_{0} + \frac{1}{4}N_{3}(0, \left(\begin{matrix} 1 & 0 &0\\
0 & 0 & 0 \\
0 & 0 & 0 \end{matrix}\right)) + \frac{1}{4}N_{3}(0, \left(\begin{matrix} 1 & 0 & 0 \\
0 & 1 & 0 \\
0 & 0 & 0 \end{matrix}\right)) + \frac{1}{4}N_{3}(0, \left(\begin{matrix} 1 & 1 & 1 \\
1 & 1 & 1 \\
1 & 1 & 1 \end{matrix}\right))
\]</span></p>
<pre class="r"><code>set.seed(1)
Sigma = cbind(c(1,0.7,0.2), c(0.7,1,0.4), c(0.2,0.4,1))
U0 = matrix(0,3,3)
U1 = matrix(0,3,3); U1[1,1] = 1
U2 = diag(3); U2[3,3] = 0
U3 = matrix(1,3,3)
data = generate_data(n=4000, p=3, V=Sigma, Utrue = list(U0=U0, U1=U1,U2=U2,U3=U3))</code></pre>
<pre class="r"><code>m.data = mash_set_data(data$Bhat, data$Shat)
m.1by1 = mash_1by1(m.data)
strong = get_significant_results(m.1by1)
U.pca = cov_pca(m.data, 3, subset = strong)
U.ed = cov_ed(m.data, U.pca, subset = strong)
U.c = cov_canonical(m.data)</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-4-1.png" width="768" style="display: block; margin: auto;" /></p>
<ol style="list-style-type: decimal">
<li>We run the algorithm in <a href="EstimateCorMax.html">estimate cor mle</a> with 3 different initial points for <span class="math inline">\(\rho\)</span> (-0.5,0,0.5). The <span class="math inline">\(\rho\)</span> in each iteration is estimated using <code>optim</code> function. The estimated correlation is</li>
</ol>
<pre class="r"><code>Vhat.mle = estimateV(m.data, c(U.c, U.ed), init_rho = c(-0.5,0,0.5), tol=1e-4, optmethod = 'mle')
Vhat.mle$V</code></pre>
<pre><code> [,1] [,2] [,3]
[1,] 1.0000000 0.7009268 0.1588733
[2,] 0.7009268 1.0000000 0.4001842
[3,] 0.1588733 0.4001842 1.0000000</code></pre>
<ol start="2" style="list-style-type: decimal">
<li>The result uses algorithm in <a href="EstimateCorMaxEM2.html">estimate cor em</a>. <span class="math inline">\(\rho\)</span> in each iteration is the root of a third degree polynomial.</li>
</ol>
<pre class="r"><code>Vhat.em = estimateV(m.data, c(U.c, U.ed), init_rho = c(-0.5,0,0.5), tol = 1e-4, optmethod = 'em2')
Vhat.em$V</code></pre>
<pre><code> [,1] [,2] [,3]
[1,] 1.0000000 0.7086927 0.1706391
[2,] 0.7086927 1.0000000 0.4187296
[3,] 0.1706391 0.4187296 1.0000000</code></pre>
<p>The running time (in sec.) for each pairwise correlation is</p>
<pre class="r"><code>table = data.frame(rbind(Vhat.mle$ttime, Vhat.em$ttime), row.names = c('mle', 'em'))
colnames(table) = c('12','13','23')
table %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:left;">
</th>
<th style="text-align:right;">
12
</th>
<th style="text-align:right;">
13
</th>
<th style="text-align:right;">
23
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:left;">
mle
</td>
<td style="text-align:right;">
286.089
</td>
<td style="text-align:right;">
208.604
</td>
<td style="text-align:right;">
70.226
</td>
</tr>
<tr>
<td style="text-align:left;">
em
</td>
<td style="text-align:right;">
167.105
</td>
<td style="text-align:right;">
63.067
</td>
<td style="text-align:right;">
51.075
</td>
</tr>
</tbody>
</table>
<p>The time is the total running time with different initial point.</p>
<ol start="3" style="list-style-type: decimal">
<li>The result uses algorithm in <a href="EstimateCorMaxEMV.html">estimate cor em V</a>.</li>
</ol>
<pre class="r"><code>Vhat.emV = estimateV(m.data, c(U.c, U.ed), init_V = list(diag(ncol(m.data$Bhat)), clusterGeneration::rcorrmatrix(3), clusterGeneration::rcorrmatrix(3)), tol = 1e-4, optmethod = 'emV')
Vhat.emV$V</code></pre>
<pre><code> [,1] [,2] [,3]
[1,] 1.0000000 0.5487183 0.2669639
[2,] 0.5487183 1.0000000 0.4654711
[3,] 0.2669639 0.4654711 1.0000000</code></pre>
<ol start="4" style="list-style-type: decimal">
<li>Using the original truncated correlation:</li>
</ol>
<pre class="r"><code>Vhat.tru = estimate_null_correlation(m.data)
Vhat.tru</code></pre>
<pre><code> [,1] [,2] [,3]
[1,] 1.0000000 0.4296283 0.1222433
[2,] 0.4296283 1.0000000 0.3324459
[3,] 0.1222433 0.3324459 1.0000000</code></pre>
<p>The truncated correlation underestimates the correlations.</p>
<ol start="5" style="list-style-type: decimal">
<li>mash 1by1</li>
</ol>
<pre class="r"><code>V.mash = cor((data$Bhat/data$Shat)[-strong,])
V.mash</code></pre>
<pre><code> [,1] [,2] [,3]
[1,] 1.0000000 0.5313446 0.2445663
[2,] 0.5313446 1.0000000 0.4490049
[3,] 0.2445663 0.4490049 1.0000000</code></pre>
<div id="error" class="section level2">
<h2>Error</h2>
<p>Checking the estimation error:</p>
<pre class="r"><code>FError = c(norm(Vhat.mle$V - Sigma, 'F'),
norm(Vhat.em$V - Sigma, 'F'),
norm(Vhat.emV$V - Sigma, 'F'),
norm(Vhat.tru - Sigma, 'F'),
norm(V.mash - Sigma, 'F'))
OpError = c(norm(Vhat.mle$V - Sigma, '2'),
norm(Vhat.em$V - Sigma, '2'),
norm(Vhat.emV$V - Sigma, '2'),
norm(Vhat.tru - Sigma, '2'),
norm(V.mash - Sigma, '2'))
table = data.frame(FError = FError, OpError = OpError, row.names = c('mle','em','emV','trunc','m.1by1'))
table %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:left;">
</th>
<th style="text-align:right;">
FError
</th>
<th style="text-align:right;">
OpError
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:left;">
mle
</td>
<td style="text-align:right;">
0.0581773
</td>
<td style="text-align:right;">
0.0411417
</td>
</tr>
<tr>
<td style="text-align:left;">
em
</td>
<td style="text-align:right;">
0.0507627
</td>
<td style="text-align:right;">
0.0391489
</td>
</tr>
<tr>
<td style="text-align:left;">
emV
</td>
<td style="text-align:right;">
0.2516219
</td>
<td style="text-align:right;">
0.1960202
</td>
</tr>
<tr>
<td style="text-align:left;">
trunc
</td>
<td style="text-align:right;">
0.4091712
</td>
<td style="text-align:right;">
0.3049974
</td>
</tr>
<tr>
<td style="text-align:left;">
m.1by1
</td>
<td style="text-align:right;">
0.2562509
</td>
<td style="text-align:right;">
0.1915171
</td>
</tr>
</tbody>
</table>
</div>
<div id="mash-log-likelihood" class="section level2">
<h2>mash log likelihood</h2>
<p>In mash model, the model with correlation from mle has larger loglikelihood.</p>
<pre class="r"><code>m.data.mle = mash_set_data(data$Bhat, data$Shat, V=Vhat.mle$V)
m.model.mle = mash(m.data.mle, c(U.c,U.ed), verbose = FALSE)</code></pre>
<pre class="r"><code>m.data.em = mash_set_data(data$Bhat, data$Shat, V=Vhat.em$V)
m.model.em = mash(m.data.em, c(U.c,U.ed), verbose = FALSE)</code></pre>
<pre class="r"><code>m.data.emV = mash_set_data(data$Bhat, data$Shat, V=Vhat.emV$V)
m.model.emV = mash(m.data.emV, c(U.c,U.ed), verbose = FALSE)</code></pre>
<pre class="r"><code>m.data.trunc = mash_set_data(data$Bhat, data$Shat, V=Vhat.tru)
m.model.trunc = mash(m.data.trunc, c(U.c,U.ed), verbose = FALSE)</code></pre>
<pre class="r"><code>m.data.1by1 = mash_set_data(data$Bhat, data$Shat, V=V.mash)
m.model.1by1 = mash(m.data.1by1, c(U.c,U.ed), verbose = FALSE)</code></pre>
<pre class="r"><code>m.data.correct = mash_set_data(data$Bhat, data$Shat, V=Sigma)
m.model.correct = mash(m.data.correct, c(U.c,U.ed), verbose = FALSE)</code></pre>
<p>The results are summarized in table:</p>
<pre class="r"><code>null.ind = which(apply(data$B,1,sum) == 0)
V.trun = c(get_loglik(m.model.trunc), length(get_significant_results(m.model.trunc)), sum(get_significant_results(m.model.trunc) %in% null.ind))
V.mle = c(get_loglik(m.model.mle), length(get_significant_results(m.model.mle)), sum(get_significant_results(m.model.mle) %in% null.ind))
V.em = c(get_loglik(m.model.em), length(get_significant_results(m.model.em)), sum(get_significant_results(m.model.em) %in% null.ind))
V.emV = c(get_loglik(m.model.emV), length(get_significant_results(m.model.emV)), sum(get_significant_results(m.model.emV) %in% null.ind))
V.1by1 = c(get_loglik(m.model.1by1), length(get_significant_results(m.model.1by1)), sum(get_significant_results(m.model.1by1) %in% null.ind))
V.correct = c(get_loglik(m.model.correct), length(get_significant_results(m.model.correct)), sum(get_significant_results(m.model.correct) %in% null.ind))
temp = cbind(V.mle, V.em, V.emV, V.trun, V.1by1, V.correct)
colnames(temp) = c('MLE','EM','EMV', 'Truncate', 'm.1by1', 'True')
row.names(temp) = c('log likelihood', '# significance', '# False positive')
temp %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:left;">
</th>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
EM
</th>
<th style="text-align:right;">
EMV
</th>
<th style="text-align:right;">
Truncate
</th>
<th style="text-align:right;">
m.1by1
</th>
<th style="text-align:right;">
True
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:left;">
log likelihood
</td>
<td style="text-align:right;">
-17917.69
</td>
<td style="text-align:right;">
-17919.23
</td>
<td style="text-align:right;">
-17945.16
</td>
<td style="text-align:right;">
-17951.46
</td>
<td style="text-align:right;">
-17943.49
</td>
<td style="text-align:right;">
-17913.58
</td>
</tr>
<tr>
<td style="text-align:left;">
# significance
</td>
<td style="text-align:right;">
146.00
</td>
<td style="text-align:right;">
149.00
</td>
<td style="text-align:right;">
82.00
</td>
<td style="text-align:right;">
85.00
</td>
<td style="text-align:right;">
73.00
</td>
<td style="text-align:right;">
149.00
</td>
</tr>
<tr>
<td style="text-align:left;">
# False positive
</td>
<td style="text-align:right;">
1.00
</td>
<td style="text-align:right;">
1.00
</td>
<td style="text-align:right;">
0.00
</td>
<td style="text-align:right;">
1.00
</td>
<td style="text-align:right;">
0.00
</td>
<td style="text-align:right;">
1.00
</td>
</tr>
</tbody>
</table>
<p>The estimated <code>pi</code> is</p>
<pre class="r"><code>par(mfrow=c(2,3))
barplot(get_estimated_pi(m.model.mle), las=2, cex.names = 0.7, main='MLE', ylim=c(0,0.8))
barplot(get_estimated_pi(m.model.em), las=2, cex.names = 0.7, main='EM', ylim=c(0,0.8))
barplot(get_estimated_pi(m.model.emV), las=2, cex.names = 0.7, main='EMV', ylim=c(0,0.8))
barplot(get_estimated_pi(m.model.trunc), las=2, cex.names = 0.7, main='Truncate', ylim=c(0,0.8))
barplot(get_estimated_pi(m.model.1by1), las=2, cex.names = 0.7, main='m.1by1', ylim=c(0,0.8))
barplot(get_estimated_pi(m.model.correct), las=2, cex.names = 0.7, main='True', ylim=c(0,0.8))</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-19-1.png" width="672" style="display: block; margin: auto;" /></p>
</div>
<div id="roc" class="section level2">
<h2>ROC</h2>
<pre class="r"><code>m.mle.seq = ROC.table(data$B, m.model.mle)
m.em.seq = ROC.table(data$B, m.model.em)
m.emV.seq = ROC.table(data$B, m.model.emV)
m.trun.seq = ROC.table(data$B, m.model.trunc)
m.1by1.seq = ROC.table(data$B, m.model.1by1)
m.correct.seq = ROC.table(data$B, m.model.correct)</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-21-1.png" width="672" style="display: block; margin: auto;" /></p>
</div>
<div id="rrmse" class="section level2">
<h2>RRMSE</h2>
<pre class="r"><code>rrmse = rbind(RRMSE(data$B, data$Bhat, list(m.model.mle, m.model.em, m.model.emV, m.model.trunc, m.model.1by1, m.model.correct)))
colnames(rrmse) = c('MLE','EM','EMV', 'Truncate','m.1by1','True')
row.names(rrmse) = 'RRMSE'
rrmse %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:left;">
</th>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
EM
</th>
<th style="text-align:right;">
EMV
</th>
<th style="text-align:right;">
Truncate
</th>
<th style="text-align:right;">
m.1by1
</th>
<th style="text-align:right;">
True
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:left;">
RRMSE
</td>
<td style="text-align:right;">
0.5277488
</td>
<td style="text-align:right;">
0.5285126
</td>
<td style="text-align:right;">
0.5437355
</td>
<td style="text-align:right;">
0.5592648
</td>
<td style="text-align:right;">
0.5442074
</td>
<td style="text-align:right;">
0.5283068
</td>
</tr>
</tbody>
</table>
<pre class="r"><code>barplot(rrmse, ylim=c(0,(1+max(rrmse))/2), names.arg = c('MLE','EM', 'EMV','Truncate','m.1by1','True'), las=2, cex.names = 0.7, main='RRMSE')</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-23-1.png" width="672" style="display: block; margin: auto;" /></p>
</div>
</div>
<div id="more-simulations" class="section level1">
<h1>More simulations</h1>
<p>I randomly generate 10 positive definite correlation matrices, V. The sample size is 4000.</p>
<p><span class="math display">\[
\hat{z}|z \sim N_{5}(z, V)
\]</span> <span class="math display">\[
z\sim\frac{1}{4}\delta_{0} + \frac{1}{4}N_{5}(0,\left(\begin{matrix} 1 & \mathbf{0}_{1\times 4} \\ \mathbf{0}_{4\times 1} & \mathbf{0}_{4\times 4} \end{matrix}\right)) + \frac{1}{4}N_{5}(0,\left(\begin{matrix} \mathbf{1}_{2\times 2} & \mathbf{0}_{1\times 3} \\ \mathbf{0}_{3\times 1} & \mathbf{0}_{3\times 3} \end{matrix}\right)) + \frac{1}{4}N_{5}(0,\mathbf{1}_{5\times 5})
\]</span></p>
<pre class="r"><code>set.seed(100)
n=4000; p = 5
U0 = matrix(0,p,p)
U1 = U0; U1[1,1] = 1
U2 = U0; U2[c(1:2), c(1:2)] = 1
U3 = matrix(1, p,p)
Utrue = list(U0 = U0, U1 = U1, U2 = U2, U3 = U3)
for(t in 1:10){
Vtrue = clusterGeneration::rcorrmatrix(p)
data = generate_data(n, p, Vtrue, Utrue)
# mash cov
m.data = mash_set_data(Bhat = data$Bhat, Shat = data$Shat)
m.1by1 = mash_1by1(m.data)
strong = get_significant_results(m.1by1)
U.pca = cov_pca(m.data, 3, subset = strong)
U.ed = cov_ed(m.data, U.pca, subset = strong)
U.c = cov_canonical(m.data)
Vhat.mle <- estimateV(m.data, c(U.c, U.ed), init_rho = c(-0.5,0,0.5), tol=1e-4, optmethod = 'mle')
Vhat.em <- estimateV(m.data, c(U.c, U.ed), init_rho = c(-0.5,0,0.5), tol=1e-4, optmethod = 'em2')
Vhat.emV <- estimateV(m.data, c(U.c, U.ed), init_V = list(diag(ncol(m.data$Bhat)), clusterGeneration::rcorrmatrix(p), clusterGeneration::rcorrmatrix(p)),tol=1e-4, optmethod = 'emV')
saveRDS(list(V.true = Vtrue, V.mle = Vhat.mle, V.em = Vhat.em, V.emV = Vhat.emV, data = data, strong=strong),
paste0('../output/MASH.result.',t,'.rds'))
}</code></pre>
<pre class="r"><code>files = dir("../output/AddEMV/"); files = files[grep("MASH.result",files)]
times = length(files)
result = vector(mode="list",length = times)
for(i in 1:times) {
result[[i]] = readRDS(paste("../output/AddEMV/", files[[i]], sep=""))
}</code></pre>
<pre class="r"><code>mle.pd = numeric(times)
em.pd = numeric(times)
for(i in 1:times){
m.data = mash_set_data(result[[i]]$data$Bhat, result[[i]]$data$Shat)
result[[i]]$V.trun = estimate_null_correlation(m.data, apply_lower_bound = FALSE)
m.1by1 = mash_1by1(m.data)
strong = get_significant_results(m.1by1)
result[[i]]$V.1by1 = cor(m.data$Bhat[-strong,])
U.c = cov_canonical(m.data)
U.pca = cov_pca(m.data, 3, subset = strong)
U.ed = cov_ed(m.data, U.pca, subset = strong)
m.data.true = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.true)
m.model.true = mash(m.data.true, c(U.c,U.ed), verbose = FALSE)
m.data.trunc = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.trun)
m.model.trunc = mash(m.data.trunc, c(U.c,U.ed), verbose = FALSE)
m.data.1by1 = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.1by1)
m.model.1by1 = mash(m.data.1by1, c(U.c,U.ed), verbose = FALSE)
m.data.emV = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.emV$V)
m.model.emV = mash(m.data.emV, c(U.c,U.ed), verbose = FALSE)
# MLE
m.model.mle = m.model.mle.F = m.model.mle.2 = list()
R <- tryCatch(chol(result[[i]]$V.mle$V),error = function (e) FALSE)
if(is.matrix(R)){
mle.pd[i] = 1
m.data.mle = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.mle$V)
m.model.mle = mash(m.data.mle, c(U.c,U.ed), verbose = FALSE)
}else{
V.mle.near.F = as.matrix(Matrix::nearPD(result[[i]]$V.mle$V, conv.norm.type = 'F', keepDiag = TRUE)$mat)
V.mle.near.2 = as.matrix(Matrix::nearPD(result[[i]]$V.mle$V, conv.norm.type = '2', keepDiag = TRUE)$mat)
result[[i]]$V.mle.F = V.mle.near.F
result[[i]]$V.mle.2 = V.mle.near.2
# mashmodel
m.data.mle.F = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = V.mle.near.F)
m.model.mle.F = mash(m.data.mle.F, c(U.c,U.ed), verbose = FALSE)
m.data.mle.2 = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = V.mle.near.2)
m.model.mle.2 = mash(m.data.mle.2, c(U.c,U.ed), verbose = FALSE)
}
# EM
m.model.em = m.model.em.F = m.model.em.2 = list()
R <- tryCatch(chol(result[[i]]$V.em$V),error = function (e) FALSE)
if(is.matrix(R)){
em.pd[i] = 1
m.data.em = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = result[[i]]$V.em$V)
m.model.em = mash(m.data.em, c(U.c,U.ed), verbose = FALSE)
}else{
V.em.near.F = as.matrix(Matrix::nearPD(result[[i]]$V.em$V, conv.norm.type = 'F', keepDiag = TRUE)$mat)
V.em.near.2 = as.matrix(Matrix::nearPD(result[[i]]$V.em$V, conv.norm.type = '2', keepDiag = TRUE)$mat)
result[[i]]$V.em.F = V.em.near.F
result[[i]]$V.em.2 = V.em.near.2
# mashmodel
m.data.em.F = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = V.em.near.F)
m.model.em.F = mash(m.data.em.F, c(U.c,U.ed), verbose = FALSE)
m.data.em.2 = mash_set_data(Bhat = m.data$Bhat, Shat = m.data$Shat, V = V.em.near.2)
m.model.em.2 = mash(m.data.em.2, c(U.c,U.ed), verbose = FALSE)
}
result[[i]]$m.model = list(m.model.true = m.model.true, m.model.trunc = m.model.trunc,
m.model.1by1 = m.model.1by1, m.model.emV = m.model.emV,
m.model.mle = m.model.mle,
m.model.mle.F = m.model.mle.F, m.model.mle.2 = m.model.mle.2,
m.model.em = m.model.em,
m.model.em.F = m.model.em.F, m.model.em.2 = m.model.em.2)
}</code></pre>
<div id="error-1" class="section level2">
<h2>Error</h2>
<p>The Frobenius norm is</p>
<pre class="r"><code>temp = matrix(0,nrow = times, ncol = 7)
for(i in 1:times){
temp[i, ] = error.cor(result[[i]], norm.type='F', mle.pd = mle.pd[i], em.pd = em.pd[i])
}
colnames(temp) = c('Trunc','m.1by1', 'MLE','MLE.F', 'EM', 'EM.F', 'EMV')
temp %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:right;">
Trunc
</th>
<th style="text-align:right;">
m.1by1
</th>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
MLE.F
</th>
<th style="text-align:right;">
EM
</th>
<th style="text-align:right;">
EM.F
</th>
<th style="text-align:right;">
EMV
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:right;">
0.5847549
</td>
<td style="text-align:right;">
0.9286016
</td>
<td style="text-align:right;">
0.3185293
</td>
<td style="text-align:right;">
0.3053278
</td>
<td style="text-align:right;">
0.0941757
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.0198618
</td>
</tr>
<tr>
<td style="text-align:right;">
0.6345196
</td>
<td style="text-align:right;">
0.8140108
</td>
<td style="text-align:right;">
0.1169154
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0981075
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8913631
</td>
</tr>
<tr>
<td style="text-align:right;">
0.7201453
</td>
<td style="text-align:right;">
0.9531300
</td>
<td style="text-align:right;">
0.2344733
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1734402
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.0216638
</td>
</tr>
<tr>
<td style="text-align:right;">
0.8370832
</td>
<td style="text-align:right;">
1.0534335
</td>
<td style="text-align:right;">
0.1968091
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.2411372
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.1115619
</td>
</tr>
<tr>
<td style="text-align:right;">
0.8206008
</td>
<td style="text-align:right;">
0.8466408
</td>
<td style="text-align:right;">
0.1194142
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1189734
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.9176794
</td>
</tr>
<tr>
<td style="text-align:right;">
0.8455747
</td>
<td style="text-align:right;">
1.1393764
</td>
<td style="text-align:right;">
0.1650726
</td>
<td style="text-align:right;">
0.1479670
</td>
<td style="text-align:right;">
0.1653665
</td>
<td style="text-align:right;">
0.1516188
</td>
<td style="text-align:right;">
1.2178752
</td>
</tr>
<tr>
<td style="text-align:right;">
0.5173056
</td>
<td style="text-align:right;">
0.8194980
</td>
<td style="text-align:right;">
0.1211599
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0861518
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8914736
</td>
</tr>
<tr>
<td style="text-align:right;">
0.8840057
</td>
<td style="text-align:right;">
0.9546138
</td>
<td style="text-align:right;">
0.1642744
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1563607
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.0169747
</td>
</tr>
<tr>
<td style="text-align:right;">
0.6535878
</td>
<td style="text-align:right;">
1.0554833
</td>
<td style="text-align:right;">
0.1240352
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1323659
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.1491978
</td>
</tr>
<tr>
<td style="text-align:right;">
0.6425639
</td>
<td style="text-align:right;">
0.9346341
</td>
<td style="text-align:right;">
0.0794776
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0700813
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
1.0196620
</td>
</tr>
</tbody>
</table>
<p>The spectral norm is</p>
<pre class="r"><code>temp = matrix(0,nrow = times, ncol = 7)
for(i in 1:times){
temp[i, ] = error.cor(result[[i]], norm.type='2', mle.pd = mle.pd[i], em.pd = em.pd[i])
}
colnames(temp) = c('Trunc','m.1by1', 'MLE','MLE.2', 'EM', 'EM.2', 'EMV')
temp %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:right;">
Trunc
</th>
<th style="text-align:right;">
m.1by1
</th>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
MLE.2
</th>
<th style="text-align:right;">
EM
</th>
<th style="text-align:right;">
EM.2
</th>
<th style="text-align:right;">
EMV
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:right;">
0.4230698
</td>
<td style="text-align:right;">
0.7526315
</td>
<td style="text-align:right;">
0.2258153
</td>
<td style="text-align:right;">
0.2205340
</td>
<td style="text-align:right;">
0.0662442
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8335745
</td>
</tr>
<tr>
<td style="text-align:right;">
0.5228925
</td>
<td style="text-align:right;">
0.6281308
</td>
<td style="text-align:right;">
0.0783098
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0677285
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.7069909
</td>
</tr>
<tr>
<td style="text-align:right;">
0.5156642
</td>
<td style="text-align:right;">
0.7945237
</td>
<td style="text-align:right;">
0.1750836
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1239438
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8577274
</td>
</tr>
<tr>
<td style="text-align:right;">
0.6529121
</td>
<td style="text-align:right;">
0.8335401
</td>
<td style="text-align:right;">
0.1484773
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1768351
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8944724
</td>
</tr>
<tr>
<td style="text-align:right;">
0.6500336
</td>
<td style="text-align:right;">
0.6377762
</td>
<td style="text-align:right;">
0.0778356
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0849870
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.7052165
</td>
</tr>
<tr>
<td style="text-align:right;">
0.5607948
</td>
<td style="text-align:right;">
0.8613851
</td>
<td style="text-align:right;">
0.1001500
</td>
<td style="text-align:right;">
0.0937816
</td>
<td style="text-align:right;">
0.1085885
</td>
<td style="text-align:right;">
0.1015186
</td>
<td style="text-align:right;">
0.9343139
</td>
</tr>
<tr>
<td style="text-align:right;">
0.3157614
</td>
<td style="text-align:right;">
0.6406310
</td>
<td style="text-align:right;">
0.0795125
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0659174
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.6982291
</td>
</tr>
<tr>
<td style="text-align:right;">
0.7134025
</td>
<td style="text-align:right;">
0.7290570
</td>
<td style="text-align:right;">
0.1090477
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.1103719
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.7987348
</td>
</tr>
<tr>
<td style="text-align:right;">
0.4767534
</td>
<td style="text-align:right;">
0.8894951
</td>
<td style="text-align:right;">
0.0964157
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0957622
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.9760343
</td>
</tr>
<tr>
<td style="text-align:right;">
0.4591215
</td>
<td style="text-align:right;">
0.7926789
</td>
<td style="text-align:right;">
0.0568252
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.0558999
</td>
<td style="text-align:right;">
0.0000000
</td>
<td style="text-align:right;">
0.8740268
</td>
</tr>
</tbody>
</table>
</div>
<div id="time" class="section level2">
<h2>Time</h2>
<p>The total running time for each matrix is</p>
<pre class="r"><code>mle.time = em.time = numeric(times)
for(i in 1:times){
mle.time[i] = sum(result[[i]]$V.mle$ttime)
em.time[i] = sum(result[[i]]$V.em$ttime)
}
temp = cbind(mle.time, em.time)
colnames(temp) = c('MLE', 'EM')
row.names(temp) = 1:10
temp %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
EM
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:right;">
3171.983
</td>
<td style="text-align:right;">
848.351
</td>
</tr>
<tr>
<td style="text-align:right;">
2341.424
</td>
<td style="text-align:right;">
704.601
</td>
</tr>
<tr>
<td style="text-align:right;">
1990.764
</td>
<td style="text-align:right;">
740.046
</td>
</tr>
<tr>
<td style="text-align:right;">
3249.162
</td>
<td style="text-align:right;">
1072.233
</td>
</tr>
<tr>
<td style="text-align:right;">
1988.220
</td>
<td style="text-align:right;">
717.808
</td>
</tr>
<tr>
<td style="text-align:right;">
2580.794
</td>
<td style="text-align:right;">
958.710
</td>
</tr>
<tr>
<td style="text-align:right;">
1928.634
</td>
<td style="text-align:right;">
597.511
</td>
</tr>
<tr>
<td style="text-align:right;">
2992.277
</td>
<td style="text-align:right;">
1114.932
</td>
</tr>
<tr>
<td style="text-align:right;">
2339.513
</td>
<td style="text-align:right;">
708.779
</td>
</tr>
<tr>
<td style="text-align:right;">
2727.560
</td>
<td style="text-align:right;">
772.780
</td>
</tr>
</tbody>
</table>
</div>
<div id="mash-log-likelihood-1" class="section level2">
<h2>mash log likelihood</h2>
<pre class="r"><code>temp = matrix(0,nrow = times, ncol = 10)
for(i in 1:times){
temp[i, ] = loglik.cor(result[[i]]$m.model, mle.pd = mle.pd[i], em.pd = em.pd[i])
}
colnames(temp) = c('True', 'Trunc','m.1by1', 'MLE','MLE.F', 'MLE.2', 'EM', 'EM.F', 'EM.2','EMV')
temp[temp == 0] = NA
temp[,-c(6,9)] %>% kable() %>% kable_styling()</code></pre>
<table class="table" style="margin-left: auto; margin-right: auto;">
<thead>
<tr>
<th style="text-align:right;">
True
</th>
<th style="text-align:right;">
Trunc
</th>
<th style="text-align:right;">
m.1by1
</th>
<th style="text-align:right;">
MLE
</th>
<th style="text-align:right;">
MLE.F
</th>
<th style="text-align:right;">
EM
</th>
<th style="text-align:right;">
EM.F
</th>
<th style="text-align:right;">
EMV
</th>
</tr>
</thead>
<tbody>
<tr>
<td style="text-align:right;">
-26039.92
</td>
<td style="text-align:right;">
-26130.50
</td>
<td style="text-align:right;">
-26112.94
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-26265.49
</td>
<td style="text-align:right;">
-26120.81
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-26136.19
</td>
</tr>
<tr>
<td style="text-align:right;">
-25669.59
</td>
<td style="text-align:right;">
-26997.92
</td>
<td style="text-align:right;">
-26950.67
</td>
<td style="text-align:right;">
-26028.96
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-25859.86
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-26967.47
</td>
</tr>
<tr>
<td style="text-align:right;">
-27473.71
</td>
<td style="text-align:right;">
-27547.11
</td>
<td style="text-align:right;">
-27535.67
</td>
<td style="text-align:right;">
-27465.76
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-27463.43
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-27551.75
</td>
</tr>
<tr>
<td style="text-align:right;">
-28215.48
</td>
<td style="text-align:right;">
-28604.98
</td>
<td style="text-align:right;">
-28646.72
</td>
<td style="text-align:right;">
-28301.41
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-28338.57
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-28669.24
</td>
</tr>
<tr>
<td style="text-align:right;">
-24988.68
</td>
<td style="text-align:right;">
-25236.18
</td>
<td style="text-align:right;">
-25110.41
</td>
<td style="text-align:right;">
-25056.59
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-25048.86
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-25123.96
</td>
</tr>
<tr>
<td style="text-align:right;">
-24299.89
</td>
<td style="text-align:right;">
-24978.79
</td>
<td style="text-align:right;">
-24972.25
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-24492.32
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-24478.09
</td>
<td style="text-align:right;">
-25020.38
</td>
</tr>
<tr>
<td style="text-align:right;">
-27574.71
</td>
<td style="text-align:right;">
-27698.40
</td>
<td style="text-align:right;">
-27662.48
</td>
<td style="text-align:right;">
-27517.65
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-27540.65
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-27684.83
</td>
</tr>
<tr>
<td style="text-align:right;">
-27941.86
</td>
<td style="text-align:right;">
-28159.65
</td>
<td style="text-align:right;">
-28182.04
</td>
<td style="text-align:right;">
-27979.99
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-27953.02
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-28222.45
</td>
</tr>
<tr>
<td style="text-align:right;">
-29788.75
</td>
<td style="text-align:right;">
-29824.45
</td>
<td style="text-align:right;">
-29921.87
</td>
<td style="text-align:right;">
-29760.81
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-29759.50
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-29954.96
</td>
</tr>
<tr>
<td style="text-align:right;">
-28542.34
</td>
<td style="text-align:right;">
-28922.58
</td>
<td style="text-align:right;">
-29163.51
</td>
<td style="text-align:right;">
-28537.96
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-28550.73
</td>
<td style="text-align:right;">
NA
</td>
<td style="text-align:right;">
-29221.61
</td>
</tr>
</tbody>
</table>
<p>The <code>NA</code> means the estimated correlation matrix is not positive definite.</p>
</div>
<div id="roc-1" class="section level2">
<h2>ROC</h2>
<pre class="r"><code>par(mfrow=c(1,2))
for(i in 1:times){
plotROC(result[[i]]$data$B, result[[i]]$m.model, mle.pd = mle.pd[i], em.pd = em.pd[i], title=paste0('Data', i, ' '))
}</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-31-1.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-31-2.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-31-3.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-31-4.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-31-5.png" width="672" style="display: block; margin: auto;" /></p>
</div>
<div id="rrmse-1" class="section level2">
<h2>RRMSE</h2>
<pre class="r"><code>par(mfrow=c(1,2))
for(i in 1:times){
rrmse = rbind(RRMSE(result[[i]]$data$B, result[[i]]$data$Bhat, result[[i]]$m.model))
barplot(rrmse, ylim=c(0,(1+max(rrmse))/2), las=2, cex.names = 0.7, main='RRMSE')
}</code></pre>
<p><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-32-1.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-32-2.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-32-3.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-32-4.png" width="672" style="display: block; margin: auto;" /><img src="figure/EstimateCorMaxMash.Rmd/unnamed-chunk-32-5.png" width="672" style="display: block; margin: auto;" /></p>
</div>
</div>
<div id="session-information" class="section level1">
<h1>Session information</h1>
<!-- Insert the session information into the document -->
<pre class="r"><code>sessionInfo()</code></pre>
<pre><code>R version 3.5.1 (2018-07-02)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS High Sierra 10.13.6
Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] kableExtra_0.9.0 knitr_1.20 mashr_0.2-11 ashr_2.2-10
loaded via a namespace (and not attached):
[1] Rcpp_0.12.18 highr_0.7
[3] compiler_3.5.1 pillar_1.3.0
[5] plyr_1.8.4 iterators_1.0.10
[7] tools_3.5.1 corrplot_0.84
[9] digest_0.6.15 viridisLite_0.3.0
[11] evaluate_0.11 tibble_1.4.2
[13] lattice_0.20-35 pkgconfig_2.0.1
[15] rlang_0.2.1 Matrix_1.2-14
[17] foreach_1.4.4 rstudioapi_0.7
[19] yaml_2.2.0 parallel_3.5.1
[21] mvtnorm_1.0-8 xml2_1.2.0
[23] httr_1.3.1 stringr_1.3.1
[25] REBayes_1.3 hms_0.4.2
[27] rprojroot_1.3-2 grid_3.5.1
[29] R6_2.2.2 rmarkdown_1.10
[31] rmeta_3.0 readr_1.1.1
[33] magrittr_1.5 scales_0.5.0
[35] backports_1.1.2 codetools_0.2-15
[37] htmltools_0.3.6 MASS_7.3-50
[39] rvest_0.3.2 assertthat_0.2.0
[41] colorspace_1.3-2 stringi_1.2.4
[43] Rmosek_8.0.69 munsell_0.5.0
[45] pscl_1.5.2 doParallel_1.0.11
[47] truncnorm_1.0-8 SQUAREM_2017.10-1
[49] clusterGeneration_1.3.4 ExtremeDeconvolution_1.3
[51] crayon_1.3.4 </code></pre>
</div>
<!-- Adjust MathJax settings so that all math formulae are shown using
TeX fonts only; see
http://docs.mathjax.org/en/latest/configuration.html. This will make
the presentation more consistent at the cost of the webpage sometimes
taking slightly longer to load. Note that this only works because the
footer is added to webpages before the MathJax javascript. -->
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
"HTML-CSS": { availableFonts: ["TeX"] }
});
</script>
<hr>
<p>
This <a href="http://rmarkdown.rstudio.com">R Markdown</a> site was created with <a href="https://github.com/jdblischak/workflowr">workflowr</a>
</p>
<hr>
<!-- To enable disqus, uncomment the section below and provide your disqus_shortname -->
<!-- disqus
<div id="disqus_thread"></div>
<script type="text/javascript">
/* * * CONFIGURATION VARIABLES: EDIT BEFORE PASTING INTO YOUR WEBPAGE * * */
var disqus_shortname = 'rmarkdown'; // required: replace example with your forum shortname
/* * * DON'T EDIT BELOW THIS LINE * * */
(function() {
var dsq = document.createElement('script'); dsq.type = 'text/javascript'; dsq.async = true;
dsq.src = '//' + disqus_shortname + '.disqus.com/embed.js';
(document.getElementsByTagName('head')[0] || document.getElementsByTagName('body')[0]).appendChild(dsq);
})();
</script>
<noscript>Please enable JavaScript to view the <a href="http://disqus.com/?ref_noscript">comments powered by Disqus.</a></noscript>
<a href="http://disqus.com" class="dsq-brlink">comments powered by <span class="logo-disqus">Disqus</span></a>
-->
</div>
</div>
</div>
<script>
// add bootstrap table styles to pandoc tables
function bootstrapStylePandocTables() {
$('tr.header').parent('thead').parent('table').addClass('table table-condensed');
}
$(document).ready(function () {
bootstrapStylePandocTables();
});
</script>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
var script = document.createElement("script");
script.type = "text/javascript";
script.src = "https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML";
document.getElementsByTagName("head")[0].appendChild(script);
})();
</script>
</body>
</html>