[Octave-bug-tracker] [bug #57136] fminsearch function diverges despite good starting point

[Giuseppe]

Follow-up Comment #2, bug #57136 (project octave):

Sure, it's the log-likeliwood function for the LEVD distribution


function loglik = loglik_LEVD(x, data)

lambda = x(1);
delta = x(2);

loglik = -sum(-log(delta) - (data - lambda)./(delta) - exp(-(data -
lambda)./delta));

end


where the data are the following


data =

   1.0e+07 *

    1.7387
    1.7877
    1.8308
    1.8312
    1.8635
    1.8804
    1.9491
    1.9602
    1.9874
    2.0843
    2.1583
    2.2554
    2.3152
    2.3533
    2.4096
    2.4172
    2.4327
    2.6184
    2.6950
    2.8036

[comment #1 comment #1:]
> Can you post the loglik_LEVD.m file so that we can attemp to reproduce this.

    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?57136>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/