1. 16 Mar, 2017 6 commits
  2. 15 Mar, 2017 4 commits
  3. 13 Mar, 2017 1 commit
  4. 10 Mar, 2017 2 commits
  5. 08 Mar, 2017 1 commit
  6. 07 Mar, 2017 2 commits
  7. 01 Mar, 2017 2 commits
  8. 24 Feb, 2017 1 commit
  9. 15 Feb, 2017 2 commits
  10. 13 Feb, 2017 1 commit
  11. 10 Feb, 2017 1 commit
  12. 09 Feb, 2017 3 commits
  13. 08 Feb, 2017 7 commits
    • Stéphane Adjemian's avatar
      Removed optimizers/fs2000_12.mod from testsuite... · 593d62a7
      Stéphane Adjemian authored
      Because the Global Optimization Toolbox is not installed on dynbot@sedna.
      593d62a7
    • Stéphane Adjemian's avatar
      Removed penalty_hessian routine. · 9fbef0c1
      Stéphane Adjemian authored
       + Code factorization.
       + Added an option for using the penalized objective when computing numerically
       the hessian at the mode.
      
      Previous behaviour (introduced with penalty_hessian routine) was to compute the
      hessian matrix at the mode with the penalized objective function (instead of
      the original objective function). This behaviour hides problematic situations,
      where the computed hessian (using the original objective) would not be full
      rank. For instance, if the estimation ends up with a parameter on (or very
      close to) the bounds of its possible values (which is often not a desirable
      outcome), the estimated posterior variance would be zero for this
      parameter (with the original objective) because the hessian is not finite in
      this direction, while the posterior variance would be positive if the penalized
      objective is used instead. But this estimate would not be reliable by
      construction of the penalty which is quite ad-hoc (more fundamentally I do not
      think that there exists any rational for approximating the covariance matrix
      with the inverse of the hessian matrix if the mode is on the border of the set
      of possible values).
      
      This commit restore the behaviour previous to 2446ab02.
      
      An option is available for computing the hessian with the penalized
      objective function.
      9fbef0c1
    • Stéphane Adjemian's avatar
      Changed priors in tests/optimizers/fs2000_* integration tests. · b14cf33c
      Stéphane Adjemian authored
      To avoid asymptote at zero on the autoregressive parameter rho.
      b14cf33c
    • Stéphane Adjemian's avatar
      Added new optimization routine (particleswarm). · 5b73d7d5
      Stéphane Adjemian authored
      Only available under Matlab if the Global Optimization Toolbox is installed.
      5b73d7d5
    • Stéphane Adjemian's avatar
      Completed header + cosmetic changes. · 35078e13
      Stéphane Adjemian authored
      35078e13
    • Stéphane Adjemian's avatar
      Cosmetic changes. · 8d4abe4a
      Stéphane Adjemian authored
      8d4abe4a
    • Stéphane Adjemian's avatar
      Added check on Octave version. · e71e89bb
      Stéphane Adjemian authored
      e71e89bb
  14. 01 Feb, 2017 2 commits
  15. 31 Jan, 2017 2 commits
  16. 27 Jan, 2017 3 commits