[ADMB Users] a difference in estimates between versions?

[Saang-Yoon]

Thank you all very much for your responses.  I have a better idea
now.   I was obsessed only by the successful calculation of Hessian
matrix.  I overlooked the maximum gradient value.   By the way, could
you tell me how to spit out the maximum gradient value through
REPORT_SECTION?   I can see it from the calculation output and also
from the PAR file, but I would like to have it via REPORT_SECTION?
Saang-Yoon

On Oct 5, 11:37 pm, Steve Martell <s.mart... at fisheries.ubc.ca> wrote:
> Here are the results from a Mac!  Based on revision 89 (July 14, 2011), last time I committed to the source code.
>
>  index   name   value      std.dev  
>      1   r    5.0287e-01 5.9674e-02
>      2   B1   8.9118e+04 1.1292e+04
>      3   K    9.0878e+04 6.0853e+03
>      4   lnq -2.1634e+00 1.7195e-01
>
> I think this might be a bit of a dog chasing its tail.  Why not use one of the benchmark models for this issue instead of this code.  If the same problem exists with a benchmark model (e.g. catage) then Saang-Yoon might have a  point, otherwise it could just be these uninformative data.
>
> By the way, it might help to rescale the data I noticed the latent variables here span 5 orders of magnitude.
>
> S
>
> On 2011-10-05, at 11:27 AM, Derek Seiple wrote:
>
>
>
>
>
> > I ran Saang-Yoon's attached example on 32-bit ubuntu with version 9.1,
> > 10.1, and the latest version with bug fixes. In ALL THREE I got the
> > following .std file.
>
> > index   name   value      std dev
> >     1   r    4.9790e-01 4.4611e-02
> >     2   B1   8.8262e+04 9.0362e+03
> >     3   K    9.1796e+04 4.5790e+03
> >     4   lnq -2.1705e+00 1.4898e-01
>
> > There was no difference in values amongst the versions. You will note
> > though that these values are slightly different from the ones he
> > supplied for ADMB IDE 450-1. This suggests to me that it is either a
> > 32-bit-64-bit difference or a windows difference.
>
> > Can anyone else run the attached model for 64-bit on linux and windows
> > or 32-bit on windows for the two versions of ADMB and report the .std
> > file for comparison?
>
> > Derek
>
> > On Wed, Oct 5, 2011 at 10:28 AM, dave fournier <da... at otter-rsch.com> wrote:
> >> It is always interesting to see what kind of posts produce the most
> >> response.
> >> certainly none of mine. Anyway
> >> there are a few standard things you can do to narrow this down.
> >> First is to take the output from each model and put it into the other
> >> version and see if the numbers
> >> stay the same. That verifies that the code produces the same function value.
> >> Also
> >> you could reduce the convergence criterion as in -crit 1.e-6 to see if you
> >> can get a little closer to the
> >> actual minimum.
>
> >> Although it is not relevant to the discussion your approximation to the
> >> differential equation
> >> for the population dynamics is gross.  See the semi-implicit version in the
> >> original
> >> manual for an idea of how to do it right.
> >> _______________________________________________
> >> Users mailing list
> >> Us... at admb-project.org
> >>http://lists.admb-project.org/mailman/listinfo/users
>
> > <gbytffall.tpl><gbytffall.dat>_____________________________________________­__
> > Users mailing list
> > Us... at admb-project.org
> >http://lists.admb-project.org/mailman/listinfo/users
>
> _______________________________________________
> Users mailing list
> Us... at admb-project.orghttp://lists.admb-project.org/mailman/listinfo/users- Hide quoted text -
>
> - Show quoted text -