<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	>

<channel>
	<title>depth first search &#187; mcmc</title>
	<atom:link href="http://www.depthfirstsearch.net/blog/tag/mcmc/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.depthfirstsearch.net/blog</link>
	<description>“We can only see a short distance ahead, but we can see plenty there that needs to be done.&#34;</description>
	<lastBuildDate>Sun, 05 Sep 2010 20:16:16 +0000</lastBuildDate>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0.1</generator>
		<item>
		<title>New Algorithm!</title>
		<link>http://www.depthfirstsearch.net/blog/2008/07/09/new-algorithm/</link>
		<comments>http://www.depthfirstsearch.net/blog/2008/07/09/new-algorithm/#comments</comments>
		<pubDate>Thu, 10 Jul 2008 03:30:22 +0000</pubDate>
		<dc:creator>JS</dc:creator>
				<category><![CDATA[admin]]></category>
		<category><![CDATA[computer science]]></category>
		<category><![CDATA[mcmc]]></category>
		<category><![CDATA[python]]></category>

		<guid isPermaLink="false">http://www.depthfirstsearch.net/blog/?p=482</guid>
		<description><![CDATA[I&#8217;ve added a new algorithm to my continuing series of Python implementations. A rather simple Metropolis-Hastings algorithm is ready for you perusal. All implementations in the series are now linked in the sidebar. Happy hacking! I&#8217;ve also updated the site license. You are now free to copy and share anything written by me on this [...]]]></description>
			<content:encoded><![CDATA[<p>I&#8217;ve added a new algorithm to my continuing series of Python implementations. A rather simple <a href="http://www.depthfirstsearch.net/blog/metropolis/">Metropolis-Hastings algorithm</a> is ready for you perusal. All implementations in the series are now linked in the sidebar. Happy hacking!</p>
<p>I&#8217;ve also updated the site license. You are now free to copy and share anything written by me on this site (including the example code) provided you abide by the <a href="http://creativecommons.org/licenses/by-nc-sa/3.0/us/">Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License</a>.</p>
<p>Now for some figures. A histogram of the resulting sample sequence (starting from x = 10 and running for 2400 steps):</p>
<p><a href='http://www.depthfirstsearch.net/blog/wp-content/uploads/2008/07/fig2.png'><img src="http://www.depthfirstsearch.net/blog/wp-content/uploads/2008/07/fig2-300x225.png" alt="" title="Metropolis-Hastings Histogram" width="300" height="225" class="aligncenter size-medium wp-image-484" /></a></p>
<p>A time series plot of the resulting samples:</p>
<p><a href='http://www.depthfirstsearch.net/blog/wp-content/uploads/2008/07/fig1.png'><img src="http://www.depthfirstsearch.net/blog/wp-content/uploads/2008/07/fig1-300x225.png" alt="" title="Sample Time Series" width="300" height="225" class="alignnone size-medium wp-image-483" /></a></p>
<p>As McKay points out, the random walk behavior is undesirable and requires that a large number of samples be taken for such a small state space.</p>
<p>UPDATE: My plots and code have an error. I don&#8217;t actually reject the rejected samples. I suspect this is why the left and right most buckets have more samples than they should.</p>
]]></content:encoded>
			<wfw:commentRss>http://www.depthfirstsearch.net/blog/2008/07/09/new-algorithm/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
	</channel>
</rss>
