<?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: Two Stage Least Squares in SAS Procedures</title>
    <link>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248236#M56397</link>
    <description>&lt;P&gt;If you had SAS/ETS, you could use &lt;A href="https://communities.sas.com/t5/SAS-Statistical-Procedures/Running-Hausman-test-to-compare-2SLS-and-OLS-models/td-p/158351" target="_self"&gt;a single call to PROC SYSLIN&lt;/A&gt;.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Although I am not an expert in this area, I believe you can call PROC REG twice to perform this analysis,&lt;/P&gt;
&lt;P&gt;based on comments in &lt;FONT color="#000000"&gt;&lt;EM&gt;&lt;FONT size="2"&gt;Using SAS for Econometrics&lt;/FONT&gt;&lt;/EM&gt;&lt;A href="http://www.principlesofeconometrics.com/poe4/usingsas.htm" target="_blank"&gt;&lt;FONT size="2"&gt;&lt;STRONG&gt;,&lt;/STRONG&gt;&lt;/FONT&gt;&lt;/A&gt;&amp;nbsp;&lt;/FONT&gt;which is the SAS supplement to Hill, Griffiths and&amp;nbsp;Lim, &lt;EM&gt;Principles of Econometrics&lt;/EM&gt;, 4th ed.&amp;nbsp; Here is a link to the &lt;A href="http://www.principlesofeconometrics.com/poe4/poe4sas_files/chap10.sas" target="_self"&gt;Chapter 10 exercises&lt;/A&gt;, which compare 2-stage PROC REG with PROC SYSLIN.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;From the Chapter 10 code, it looks like you use PROC REG to generate&amp;nbsp;residuals for the first stage, and you call PROC REG again and use those residuals in the second stage regression to get Hausman's test.&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Fri, 05 Feb 2016 14:08:49 GMT</pubDate>
    <dc:creator>Rick_SAS</dc:creator>
    <dc:date>2016-02-05T14:08:49Z</dc:date>
    <item>
      <title>Two Stage Least Squares</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248130#M56373</link>
      <description>&lt;P&gt;If you have SAS/STAT but not SAS/ETS, is there a proc available to do two stage least squares? Also is there a good way to conduct a Hausman test in SAS/STAT? Thanks!&lt;/P&gt;</description>
      <pubDate>Thu, 04 Feb 2016 23:58:01 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248130#M56373</guid>
      <dc:creator>bstevenson</dc:creator>
      <dc:date>2016-02-04T23:58:01Z</dc:date>
    </item>
    <item>
      <title>Re: Two Stage Least Squares</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248166#M56379</link>
      <description>&lt;P&gt;You could try using R... &lt;A href="http://novicemetrics.blogspot.co.uk/2010/05/two-stage-least-squares.html" target="_blank"&gt;http://novicemetrics.blogspot.co.uk/2010/05/two-stage-least-squares.html&lt;/A&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Norman.&lt;/P&gt;</description>
      <pubDate>Fri, 05 Feb 2016 05:56:13 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248166#M56379</guid>
      <dc:creator>Norman21</dc:creator>
      <dc:date>2016-02-05T05:56:13Z</dc:date>
    </item>
    <item>
      <title>Re: Two Stage Least Squares</title>
      <link>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248236#M56397</link>
      <description>&lt;P&gt;If you had SAS/ETS, you could use &lt;A href="https://communities.sas.com/t5/SAS-Statistical-Procedures/Running-Hausman-test-to-compare-2SLS-and-OLS-models/td-p/158351" target="_self"&gt;a single call to PROC SYSLIN&lt;/A&gt;.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Although I am not an expert in this area, I believe you can call PROC REG twice to perform this analysis,&lt;/P&gt;
&lt;P&gt;based on comments in &lt;FONT color="#000000"&gt;&lt;EM&gt;&lt;FONT size="2"&gt;Using SAS for Econometrics&lt;/FONT&gt;&lt;/EM&gt;&lt;A href="http://www.principlesofeconometrics.com/poe4/usingsas.htm" target="_blank"&gt;&lt;FONT size="2"&gt;&lt;STRONG&gt;,&lt;/STRONG&gt;&lt;/FONT&gt;&lt;/A&gt;&amp;nbsp;&lt;/FONT&gt;which is the SAS supplement to Hill, Griffiths and&amp;nbsp;Lim, &lt;EM&gt;Principles of Econometrics&lt;/EM&gt;, 4th ed.&amp;nbsp; Here is a link to the &lt;A href="http://www.principlesofeconometrics.com/poe4/poe4sas_files/chap10.sas" target="_self"&gt;Chapter 10 exercises&lt;/A&gt;, which compare 2-stage PROC REG with PROC SYSLIN.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;From the Chapter 10 code, it looks like you use PROC REG to generate&amp;nbsp;residuals for the first stage, and you call PROC REG again and use those residuals in the second stage regression to get Hausman's test.&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Fri, 05 Feb 2016 14:08:49 GMT</pubDate>
      <guid>https://communities.sas.com/t5/SAS-Procedures/Two-Stage-Least-Squares/m-p/248236#M56397</guid>
      <dc:creator>Rick_SAS</dc:creator>
      <dc:date>2016-02-05T14:08:49Z</dc:date>
    </item>
  </channel>
</rss>

