<?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>Aptech &#187; Answers for "How can I avoid catastrophic failure of precision in Hessian calculations?"</title>
	<atom:link href="http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.aptech.com</link>
	<description></description>
	<lastBuildDate>Fri, 08 Feb 2013 19:12:28 +0000</lastBuildDate>
	<language>en-US</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.4.1</generator>
		<item>
		<title>By: admin</title>
		<link>http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/#answer-1274</link>
		<comments>http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/#answer-1274#comments</comments>
		<pubDate>Fri, 31 Aug 2012 08:56:22 +0000</pubDate>
		<dc:creator>admin</dc:creator>
		
		<guid isPermaLink="false">http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/#answer-1274</guid>
		<description><![CDATA[There are a number of possible solutions to these problems: 1. Increasing Precision.  The first task is to work to remove large numbers from the calculations in the procedure computing the log-likelihood or objective function.  Often all that’s needed is to &#8230; <a href="http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/#answer-1274">Continue reading <span class="meta-nav">&#8594;</span></a>]]></description>
			<content:encoded><![CDATA[<p>There are a number of possible solutions to these problems:</p>
<p>1. <strong>Increasing Precision</strong>.  The first task is to work to remove large numbers from the calculations in the procedure computing the log-likelihood or objective function.  Often all that’s needed is to scale the data.  After that inspect the procedures to see of any operations can be simplified to prevent a loss of precision.  For example, disaggregate the calculation.  Instead of</p>
<pre> a(b + c)</pre>
<p>try this</p>
<pre> ab + ac</pre>
<p>2. <strong>Using</strong> <strong>Analytical Derivatives.</strong>  Analytical derivatives avoid the problems associated with numerical derivatives.  The latest versions of <strong>Constrained Optimization MT (COMT)</strong> and <strong>Constrained Maximum Likelihood MT (CMLMT)</strong> do not require calculating all of the derivatives analytically.  You can pick out the ones with the greatest potential for loss of precision and the remaining will be computed numerically.  Also, when analytical derivatives are available, the program will compute the Hessian using the analytical derivatives increasing their precision as well.</p>
]]></content:encoded>
			<wfw:commentRss>http://www.aptech.com/questions/how-can-i-avoid-catastrophic-failure-of-precision-in-hessian-calculations/#answer-1274/feed/</wfw:commentRss>
		<slash:comments>0</slash:comments>
		</item>
	</channel>
</rss>

<!-- Performance optimized by W3 Total Cache. Learn more: http://www.w3-edge.com/wordpress-plugins/

Served from: www.aptech.com @ 2013-02-09 01:37:33 --