<?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 Nested linear combination of variables in SAS Procedures</title>
    <link>https://communities.sas.com/t5/SAS-Procedures/Nested-linear-combination-of-variables/m-p/32347#M7780</link>
    <description>Hi all, I would appreciate your help on the following problem.&lt;BR /&gt;
&lt;BR /&gt;
I need to estimate a choice model (logit) and one independent variable will show by itself as well as a combination of five other variables. A simple example would be:&lt;BR /&gt;
&lt;BR /&gt;
choice=intercept+price+price*demographics+error&lt;BR /&gt;
demographics=income+distance&lt;BR /&gt;
&lt;BR /&gt;
Now is there a way I can nest/embed the linear combination into the model statement so I do not have to calculate the estimates of income and distance manually?&lt;BR /&gt;
Of course my model has many many variables so I want to find a way to let SAS spit those out for me.&lt;BR /&gt;
&lt;BR /&gt;
Thank you very very much.</description>
    <pubDate>Thu, 17 Mar 2011 19:29:35 GMT</pubDate>
    <dc:creator>elvispriscilla</dc:creator>
    <dc:date>2011-03-17T19:29:35Z</dc:date>
    <item>
      <title>Nested linear combination of variables</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Nested-linear-combination-of-variables/m-p/32347#M7780</link>
      <description>Hi all, I would appreciate your help on the following problem.&lt;BR /&gt;
&lt;BR /&gt;
I need to estimate a choice model (logit) and one independent variable will show by itself as well as a combination of five other variables. A simple example would be:&lt;BR /&gt;
&lt;BR /&gt;
choice=intercept+price+price*demographics+error&lt;BR /&gt;
demographics=income+distance&lt;BR /&gt;
&lt;BR /&gt;
Now is there a way I can nest/embed the linear combination into the model statement so I do not have to calculate the estimates of income and distance manually?&lt;BR /&gt;
Of course my model has many many variables so I want to find a way to let SAS spit those out for me.&lt;BR /&gt;
&lt;BR /&gt;
Thank you very very much.</description>
      <pubDate>Thu, 17 Mar 2011 19:29:35 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Nested-linear-combination-of-variables/m-p/32347#M7780</guid>
      <dc:creator>elvispriscilla</dc:creator>
      <dc:date>2011-03-17T19:29:35Z</dc:date>
    </item>
    <item>
      <title>Re: Nested linear combination of variables</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Nested-linear-combination-of-variables/m-p/32348#M7781</link>
      <description>Hi,&lt;BR /&gt;
&lt;BR /&gt;
What I understood from your description is that you want to use a linear combination of income and distance as a single variable (demographics) in the logistic regression model for choice. &lt;BR /&gt;
&lt;BR /&gt;
I think you could use the variance components analysis to create your demographics variable. Then it can be plugged into your choice model. With ODS and a bit of macro programming this can be automated.</description>
      <pubDate>Fri, 18 Mar 2011 15:45:41 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Nested-linear-combination-of-variables/m-p/32348#M7781</guid>
      <dc:creator>statsplank</dc:creator>
      <dc:date>2011-03-18T15:45:41Z</dc:date>
    </item>
  </channel>
</rss>

