<html><head></head><body>A Bayesian approach to regression with inequality
constraints on coefficients is to first estimate the regression without constraints,
then sample from the distribution of the coefficients, discarding all draws
that violate the constraints, and finally calculate summary statistics from
the subsample that is consistent with the constraints. Andrew Gelman et
al. explain how to do that in their Bayesian Data Analysis. I believe they
implemented their procedures in S-Plus. If any one has written similar programs
in R, I would like very much to hear them.<br>
<br>
Adam Gehr wrote:<br>
<blockquote type="cite" cite="mid:3A551C37.3EBFDE95@mozart.fin.depaul.edu"><pre wrap=""><br>"Strumila, John" wrote:<br></pre>
<blockquote type="cite"><pre wrap="">gday R gurus,<br><br>I have a multivariate regression for which I want to constrain the<br>coefficients to be > 0. Is this possible?<br><br>I've check the doco and searched CRAN but can't find anything.<br><br>thanks,<br>John Strumila<br></pre></blockquote>
<pre wrap=""><!----><br>I've been doing something like that (regression coefficients<br>constrained <br>to be > 0 and also forced to sum to 1) using the quadratic<br>programming program solve.QP in package quadprog. <br><br> Adam Gehr<br>-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-..-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-<br>r-help mailing list -- Read <a class="moz-txt-link-freetext" href="http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html">http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html</a><br>Send "info", "help", or "[un]subscribe"<br>(in the "body", not the subject !) To: <a class="moz-txt-link-abbreviated" href="mailto:r-help-request@stat.math.ethz.ch">r-help-request@stat.math.ethz.ch</a><br>_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._<br><br><br></pre>
</blockquote>
<br>
<pre class="moz-signature">--
John P. Burkett
Department of Economics
University of Rhode Island
10 Chafee Road, Suite 3
Kingston, RI 02881-0808
phone (401) 874-4122
fax (401) 874-2858</pre>
<br>
</body></html>