<?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: Calculating Trend Indicators</title>
	<atom:link href="http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/</link>
	<description>Thoughts, musings, and, hopefully, not too many redundancies on the world of business data. If you missed the irony in the previous sentence, you may struggle with my writing style.</description>
	<lastBuildDate>Tue, 07 Feb 2012 18:26:03 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: John</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-32295</link>
		<dc:creator>John</dc:creator>
		<pubDate>Wed, 18 Nov 2009 21:07:27 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-32295</guid>
		<description>How many data points are needed to prove a trend?</description>
		<content:encoded><![CDATA[<p>How many data points are needed to prove a trend?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: An Excel Dashboard Widget &#124; Gilligan on Data by Tim Wilson</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-31649</link>
		<dc:creator>An Excel Dashboard Widget &#124; Gilligan on Data by Tim Wilson</dc:creator>
		<pubDate>Mon, 16 Nov 2009 14:03:15 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-31649</guid>
		<description>[...] out Excel-based dashboard structures and processes of late. I also wrote a few weeks ago about calculating trend indicators. A natural follow-on to both of those posts is a look at the &#8220;metric widget&#8221; that I use [...]</description>
		<content:encoded><![CDATA[<p>[...] out Excel-based dashboard structures and processes of late. I also wrote a few weeks ago about calculating trend indicators. A natural follow-on to both of those posts is a look at the &#8220;metric widget&#8221; that I use [...]</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Patrick McMahon</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-28707</link>
		<dc:creator>Patrick McMahon</dc:creator>
		<pubDate>Tue, 27 Oct 2009 00:39:47 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-28707</guid>
		<description>Tim - came across this article while musing the same topic, myself.  Can you share a sample excel or some formulas to further illustrate how you are using slope and intercept within your formulas?  Thanks!</description>
		<content:encoded><![CDATA[<p>Tim &#8211; came across this article while musing the same topic, myself.  Can you share a sample excel or some formulas to further illustrate how you are using slope and intercept within your formulas?  Thanks!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: JT Buser</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-28376</link>
		<dc:creator>JT Buser</dc:creator>
		<pubDate>Wed, 21 Oct 2009 15:20:32 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-28376</guid>
		<description>Hey Tim,

Great post.  I will actually use this today when discussing data visualization with my team members.  I think bottom line, is your situation will dictate how and what statistical technique you use to trend.  This is probably why MR did not work for you.  I think it has some applications for us, but to be honest, I never even used it that much in Quality Control charts.  You may want to check out an exponentially weighted moving average.  Its like a moving average, but it gives more weight to the most recent data point and less weight to previous data points.  I have used it in the past to detect small shifts within my mean, which is typically what you want to know....</description>
		<content:encoded><![CDATA[<p>Hey Tim,</p>
<p>Great post.  I will actually use this today when discussing data visualization with my team members.  I think bottom line, is your situation will dictate how and what statistical technique you use to trend.  This is probably why MR did not work for you.  I think it has some applications for us, but to be honest, I never even used it that much in Quality Control charts.  You may want to check out an exponentially weighted moving average.  Its like a moving average, but it gives more weight to the most recent data point and less weight to previous data points.  I have used it in the past to detect small shifts within my mean, which is typically what you want to know&#8230;.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tim Wilson</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-27514</link>
		<dc:creator>Tim Wilson</dc:creator>
		<pubDate>Tue, 06 Oct 2009 11:40:52 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-27514</guid>
		<description>Excellent point, Rick! The analog in marketing would be any sort of &quot;big splash&quot; campaign that affects what you&#039;re measuring. So, then the question is: do you try to remove those outliers from the trend, or do you go deeper on the statistical front and start truly breaking out rational subgroups and building a more rigorous regression? The situation will dictate, right?</description>
		<content:encoded><![CDATA[<p>Excellent point, Rick! The analog in marketing would be any sort of &#8220;big splash&#8221; campaign that affects what you&#8217;re measuring. So, then the question is: do you try to remove those outliers from the trend, or do you go deeper on the statistical front and start truly breaking out rational subgroups and building a more rigorous regression? The situation will dictate, right?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Rick</title>
		<link>http://www.gilliganondata.com/index.php/2009/10/05/calculating-trend-indicators/comment-page-1/#comment-27468</link>
		<dc:creator>Rick</dc:creator>
		<pubDate>Mon, 05 Oct 2009 16:31:07 +0000</pubDate>
		<guid isPermaLink="false">http://www.gilliganondata.com/?p=548#comment-27468</guid>
		<description>This is a great discussion of the issues of trending.  I believe their is one concept that needs to be included.  Trending is generally used to forecast future performance.  Your March Madness example is perfect.  Trendlines only are indicative of the future if you assume that nothing changes either externally or internally in your processes.</description>
		<content:encoded><![CDATA[<p>This is a great discussion of the issues of trending.  I believe their is one concept that needs to be included.  Trending is generally used to forecast future performance.  Your March Madness example is perfect.  Trendlines only are indicative of the future if you assume that nothing changes either externally or internally in your processes.</p>
]]></content:encoded>
	</item>
</channel>
</rss>

