<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	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/"
		>
<channel>
	<title>Comments on: The Top Scores for Canabalt, Take 2</title>
	<atom:link href="http://www.johnmyleswhite.com/notebook/2009/11/15/the-top-scores-for-canabalt-take-2/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.johnmyleswhite.com/notebook/2009/11/15/the-top-scores-for-canabalt-take-2/</link>
	<description>&#34;He who refuses to do arithmetic is doomed to talk nonsense.&#34;</description>
	<lastBuildDate>Mon, 05 Jul 2010 19:57:28 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.0</generator>
	<item>
		<title>By: John Myles White</title>
		<link>http://www.johnmyleswhite.com/notebook/2009/11/15/the-top-scores-for-canabalt-take-2/comment-page-1/#comment-17789</link>
		<dc:creator>John Myles White</dc:creator>
		<pubDate>Thu, 19 Nov 2009 20:39:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.johnmyleswhite.com/?p=3605#comment-17789</guid>
		<description>Thanks for the suggestion, Owe. I had thought at one point that the distribution looked a bit like some parametrizations of the F distribution, but couldn&#039;t come up with a theoretical motivation for using it: the gamma seems much more appropriate because of its origin in a sum of exponentials. Any suggestions on how to fit the gamma in R? My first pass, below, crashes because &lt;code&gt;Nan&lt;/code&gt;&#039;s are produced.

&lt;code&gt;
library(MASS)
fitdistr(data.set$score, &quot;gamma&quot;)
&lt;/code&gt;</description>
		<content:encoded><![CDATA[<p>Thanks for the suggestion, Owe. I had thought at one point that the distribution looked a bit like some parametrizations of the F distribution, but couldn&#8217;t come up with a theoretical motivation for using it: the gamma seems much more appropriate because of its origin in a sum of exponentials. Any suggestions on how to fit the gamma in R? My first pass, below, crashes because <code>Nan</code>&#8216;s are produced.</p>
<p><code><br />
library(MASS)<br />
fitdistr(data.set$score, "gamma")<br />
</code></p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Owe Jessen</title>
		<link>http://www.johnmyleswhite.com/notebook/2009/11/15/the-top-scores-for-canabalt-take-2/comment-page-1/#comment-17788</link>
		<dc:creator>Owe Jessen</dc:creator>
		<pubDate>Thu, 19 Nov 2009 19:10:56 +0000</pubDate>
		<guid isPermaLink="false">http://www.johnmyleswhite.com/?p=3605#comment-17788</guid>
		<description>How about a gamma distribution? It should give you a fairly decent approximation.Bound at zero, and you can fit a lot of curves to it, see http://en.wikipedia.org/wiki/File:Gamma_distribution_pdf.png</description>
		<content:encoded><![CDATA[<p>How about a gamma distribution? It should give you a fairly decent approximation.Bound at zero, and you can fit a lot of curves to it, see <a href="http://en.wikipedia.org/wiki/File:Gamma_distribution_pdf.png" rel="nofollow">http://en.wikipedia.org/wiki/File:Gamma_distribution_pdf.png</a></p>
]]></content:encoded>
	</item>
</channel>
</rss>

<!-- Dynamic Page Served (once) in 1.773 seconds -->
