Hi Edgar and Jim,<div><br></div><div>Thanks for the simple example, Jim. I have never tried -mcr, but it is nice to finally give it a test. I added to your example a little bit just to verify in my mind that you do not need to re-run the minimization, but can start directly at the MCMC. My appended model is:</div>
<div><br></div><div><div>DATA_SECTION</div><div><br></div><div>PARAMETER_SECTION</div><div> init_bounded_number a(-1,2)</div><div> sdreport_number b</div><div> objective_function_value f</div><div>PROCEDURE_SECTION</div>
<div> b=a;</div><div> f = square(a);</div><div> if(mceval_phase()) {</div><div> cout << "a=" << a << endl;</div><div> }</div><div><br></div><div>I then ran </div><div>program -mcmc 1000 -mcsave 100 -mcseed 3</div>
<div>program -mceval</div><div><br></div><div>to get:</div><div><div>a=0.600564</div><div>a=0.155787</div><div>a=0.722433</div><div>a=-0.300787</div><div>a=-0.281474</div><div>a=0.272127</div><div>a=0.238472</div><div>a=0.15326</div>
<div>a=0.283672</div><div>a=0.118446</div></div><div><br></div><div><br></div><div>I then ran:</div><div>program -mcmc 1000 -mcsave 100 -mcr -noest</div><div>program -mceval</div><div><br></div><div>and got:</div><div><div>
a=0.600564</div><div>a=0.155787</div><div>a=0.722433</div><div>a=-0.300787</div><div>a=-0.281474</div><div>a=0.272127</div><div>a=0.238472</div><div>a=0.15326</div><div>a=0.283672</div><div>a=0.118446</div><div>a=-0.211523</div>
<div>a=-0.321766</div><div>a=1.74803</div><div>a=0.110279</div><div>a=-0.569084</div><div>a=0.299039</div><div>a=-0.118126</div><div>a=1.33894</div><div>a=0.0298129</div><div>a=-0.135164</div></div><div><br></div><div><br>
</div><div>A couple of things I learned are:</div><div>1) When using the -mcr option, you do not need to minimize again (use -noest to save some time)</div><div> a) and you do not need to specify any starting values using -ainp or -binp (because it starts from the previous chain)</div>
<div>2) When using -mcseed, the first value is not the MLE estimate (when not using a seed the first sample is the estimate from minimization)</div><div><br></div><div>Thanks again for spurring me to investigate this further and answer some questions that I had. Please correct me if any of my conclusions are wrong.</div>
<div>Allan</div><div><br></div><div><br></div><br><div class="gmail_quote">On Tue, Aug 7, 2012 at 5:17 PM, Jim Ianelli <span dir="ltr"><<a href="mailto:jim.ianelli@noaa.gov" target="_blank">jim.ianelli@noaa.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<div><br></div><div><div>as a minimalist test I made the model:</div><div><div>DATA_SECTION</div><div><br></div><div>
PARAMETER_SECTION</div><div> init_number a</div><div> sdreport_number b</div><div> objective_function_value f</div>
<div>PROCEDURE_SECTION</div><div> b=a;</div><div> f = square(a);</div></div><div><br></div><div>and invoked mcmc as:</div><div><br></div><div>program -mcmc 1000 -mcsave 10</div><div><br></div><div>and using readbin:</div>
<div>readbin program.psv</div><div><br></div><div>I found there was a chain of 100 elements.</div><div><br></div><div>restarting as:</div><div><br></div><div>program -mcr -mcmc 1000 -mcsave</div><div><br></div><div>I got 200 elements in the chain...</div>
<div><br></div><div>So it should work just to add -mcr with the previous arguments.</div><div><br></div><div>8 days to get 150000 MCMCs...wow.</div><div><br></div><div>Cheers,</div><div>Jim</div><div><br></div><br><div class="gmail_quote">
<div><div class="h5">
On Tue, Aug 7, 2012 at 4:58 PM, Edgar Gonzalez <span dir="ltr"><<a href="mailto:edgarjgonzalez@ymail.com" target="_blank">edgarjgonzalez@ymail.com</a>></span> wrote:<br></div></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div><div class="h5">
<div><div style="font-size:12pt;font-family:times new roman,new york,times,serif"><div style="font-family:'times new roman','new york',times,serif;font-size:12pt"><span>Hi everyone,</span></div><div style="font-family:'times new roman','new york',times,serif;font-size:12pt">
<span><br></span></div><div style="font-family:'times new roman','new york',times,serif;font-size:12pt">Does anyone know how the -mcr command works? I've searched the web and couldn't find its use, apart from the vague:</div>
<div style="font-family:'times new roman','new york',times,serif;font-size:12pt"><br></div><div><blockquote style="font-family:'times new roman','new york',times,serif;font-size:12pt;margin:0px 0px 0px 40px;border:none;padding:0px">
<div>
<div>-mcr resume previous mcmc</div><div><br></div></div></blockquote>I ran my code (which is not a random effects model) as:</div><div><br></div><div><blockquote style="margin:0 0 0 40px;border:none;padding:0px"><div>program -mcmc 500000 -mcsave -1500</div>
<div><br></div></blockquote>which took about 8 days to run. My problem is that I can only run codes for 10 days and then they're automatically stopped. So I'll have to partition my mcmc run as I require 16 million steps (calculated from a previous diagnostics run).</div>
<div><br></div><div>Could anyone help me?</div><div><br></div><div>Thanks in advance,</div><div style="font-family:'times new roman','new york',times,serif;font-size:12pt"> </div><div style="font-family:'times new roman','new york',times,serif;font-size:12pt">
Edgar J. González<br>Ecology Department<br>Science Faculty, UNAM</div></div></div><br></div></div>_______________________________________________<br>
Users mailing list<br>
<a href="mailto:Users@admb-project.org" target="_blank">Users@admb-project.org</a><br>
<a href="http://lists.admb-project.org/mailman/listinfo/users" target="_blank">http://lists.admb-project.org/mailman/listinfo/users</a><br>
<br></blockquote></div><span class="HOEnZb"><font color="#888888"><br><br clear="all"><div><br></div>-- <br><p><a name="1390399af8c54e24_SafeHtmlFilter_SafeHtmlFilter__MailAutoSig">James
Ianelli<br></a>REFM Division<br>Alaska Fisheries Science Center<br>NMFS/NOAA Building 4<br>7600 Sand Pt Way NE<br>Seattle WA 98115</p><p></p>
<p> 206 526 6510</p><p></p><br>
</font></span></div>
<br>_______________________________________________<br>
Users mailing list<br>
<a href="mailto:Users@admb-project.org">Users@admb-project.org</a><br>
<a href="http://lists.admb-project.org/mailman/listinfo/users" target="_blank">http://lists.admb-project.org/mailman/listinfo/users</a><br>
<br></blockquote></div><br></div>