<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Gamma model with log link in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215312#M11648</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Because Gamma they are all positive value , use LOG() to map it into (- infinite ,+ infinite) .&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Sat, 01 Aug 2015 13:24:51 GMT</pubDate>
    <dc:creator>Ksharp</dc:creator>
    <dc:date>2015-08-01T13:24:51Z</dc:date>
    <item>
      <title>Gamma model with log link</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215311#M11647</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Dear all.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Wish you all well.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;My apoloy that this's not a SAS question.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Can someone please englighten me on why log link but not others is used for Gamma models? What's the underlying theory?&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Thank you very much. &lt;/SPAN&gt;&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Sat, 01 Aug 2015 11:19:26 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215311#M11647</guid>
      <dc:creator>Miracle</dc:creator>
      <dc:date>2015-08-01T11:19:26Z</dc:date>
    </item>
    <item>
      <title>Re: Gamma model with log link</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215312#M11648</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Because Gamma they are all positive value , use LOG() to map it into (- infinite ,+ infinite) .&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Sat, 01 Aug 2015 13:24:51 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215312#M11648</guid>
      <dc:creator>Ksharp</dc:creator>
      <dc:date>2015-08-01T13:24:51Z</dc:date>
    </item>
    <item>
      <title>Re: Gamma model with log link</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215313#M11649</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Actually, the canonical link is the reciprocal and this is, in fact, the default link in PROC GENMOD when you specify the DIST=GAMMA option without the LINK= option.&amp;nbsp; But the log link is the one most commonly used.&amp;nbsp; The canonical link is the link function that naturally occurs when the distribution is written in the general exponential family form that is shown in the "Details: Generalized Linear Models Theory" section of the GENMOD or GLIMMIX documentation.&amp;nbsp; More explanation can be found in a reference text like McCullagh and Nelder's Generalized Linear Models book.&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Sat, 01 Aug 2015 15:36:39 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215313#M11649</guid>
      <dc:creator>StatDave</dc:creator>
      <dc:date>2015-08-01T15:36:39Z</dc:date>
    </item>
    <item>
      <title>Re: Gamma model with log link</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215314#M11650</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Dear Xia and Dave,&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;Thank you both for your reply.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="font-family: 'times new roman', times; font-size: 12pt;"&gt;I think I need to do some further readings on it.&lt;/SPAN&gt;&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Thu, 06 Aug 2015 01:16:00 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Gamma-model-with-log-link/m-p/215314#M11650</guid>
      <dc:creator>Miracle</dc:creator>
      <dc:date>2015-08-06T01:16:00Z</dc:date>
    </item>
  </channel>
</rss>

