<?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: hpgenselect effect hierarchy in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/351059#M18434</link>
    <description>&lt;P&gt;It looks like HPGENSELECT does not support changing the default hierarchy option. If you have a linear model, you can use the &lt;A href="http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_glmselect_syntax07.htm" target="_self"&gt;HIERARCHY= option in PROC GLMSELECT.&lt;/A&gt;&amp;nbsp; However, for HPGENSELECT, it looks like STEPWISE method always uses 'none' and the BACKWARD method always uses 'single.' &amp;nbsp;(For a logistic model, PROC LOGISTIC also supports changing the hierarchy option.)&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;For what it's worth, the defaults are probably what you want. In particular, you would probably NOT want to change the BACKWARD method away from the default.It's not usually a good idea to remove main effects and keep the interaction effects.&lt;/P&gt;</description>
    <pubDate>Tue, 18 Apr 2017 21:07:31 GMT</pubDate>
    <dc:creator>Rick_SAS</dc:creator>
    <dc:date>2017-04-18T21:07:31Z</dc:date>
    <item>
      <title>hpgenselect effect hierarchy</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/350764#M18432</link>
      <description>&lt;P&gt;When I run proc hpgenselect using the stepwise selection method I see that "effect hierarchy enforced" is none in the HTML output. I run it against the same data using backward selection method and get that "effect hierarchy enforced" is "single". Is there a way to set this parameter? I've googled this question and read the documentation and haven't been able to find the answer. Let me know!&lt;/P&gt;&lt;BR /&gt;&lt;IMG src="https://communities.sas.com/t5/image/serverpage/image-id/13722i938A53DA51E87EBD/image-size/large?v=1.0&amp;amp;px=600" border="0" alt="backward selection method.JPG" title="backward selection method.JPG" /&gt;</description>
      <pubDate>Tue, 18 Apr 2017 10:08:10 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/350764#M18432</guid>
      <dc:creator>jessicaking</dc:creator>
      <dc:date>2017-04-18T10:08:10Z</dc:date>
    </item>
    <item>
      <title>Re: hpgenselect effect hierarchy</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/350863#M18433</link>
      <description>&lt;P&gt;Please&amp;nbsp;post the code you are submitting.&lt;/P&gt;</description>
      <pubDate>Tue, 18 Apr 2017 14:04:38 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/350863#M18433</guid>
      <dc:creator>Rick_SAS</dc:creator>
      <dc:date>2017-04-18T14:04:38Z</dc:date>
    </item>
    <item>
      <title>Re: hpgenselect effect hierarchy</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/351059#M18434</link>
      <description>&lt;P&gt;It looks like HPGENSELECT does not support changing the default hierarchy option. If you have a linear model, you can use the &lt;A href="http://support.sas.com/documentation/cdl/en/statug/68162/HTML/default/viewer.htm#statug_glmselect_syntax07.htm" target="_self"&gt;HIERARCHY= option in PROC GLMSELECT.&lt;/A&gt;&amp;nbsp; However, for HPGENSELECT, it looks like STEPWISE method always uses 'none' and the BACKWARD method always uses 'single.' &amp;nbsp;(For a logistic model, PROC LOGISTIC also supports changing the hierarchy option.)&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;For what it's worth, the defaults are probably what you want. In particular, you would probably NOT want to change the BACKWARD method away from the default.It's not usually a good idea to remove main effects and keep the interaction effects.&lt;/P&gt;</description>
      <pubDate>Tue, 18 Apr 2017 21:07:31 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/hpgenselect-effect-hierarchy/m-p/351059#M18434</guid>
      <dc:creator>Rick_SAS</dc:creator>
      <dc:date>2017-04-18T21:07:31Z</dc:date>
    </item>
  </channel>
</rss>

