diff options
author | Johannes Ranke <jranke@uni-bremen.de> | 2019-04-10 10:17:35 +0200 |
---|---|---|
committer | Johannes Ranke <jranke@uni-bremen.de> | 2019-04-10 10:17:35 +0200 |
commit | 194659fcaccdd1ee37851725b8c72e99daa3a8cf (patch) | |
tree | edbbebe8956000b9eb725ca425b91e051571ec02 /NEWS.md | |
parent | 5814be02f286ce96d6cff8d698aea6844e4025f1 (diff) |
Adapt tests, vignettes and examples
- Write the NEWS
- Static documentation rebuilt by pkgdown
- Adapt mkinerrmin
- Fix (hopefully all) remaining problems in mkinfit
Diffstat (limited to 'NEWS.md')
-rw-r--r-- | NEWS.md | 17 |
1 files changed, 17 insertions, 0 deletions
@@ -1,3 +1,20 @@ +# mkin 0.9.49.4 (2019-04-09) +- Direct minimization of the negative log-likelihood for non-constant error models (two-component and variance by variable). In the case the error model is constant variance, least squares is used as this is more stable + +- The argument 'reweight.method' to mkinfit and mmkin is now obsolete, use 'error_model' instead + +- Remove dependency on FME, only use nlminb for optimisation + +- Use the numDeriv package to calculate hessians + +- Add a benchmark vignette to document the impact on performance. For very simple fits, the new code is a bit slower, presumably because of the time it takes to calculate the hessian matrices with and without parameter transformation + +- The code for manual weighting was removed + +- The fitting time reported in the summary now includes the calculation of the hessians + +- Adapt tests + # mkin 0.9.48.1 (2019-03-04) - Add the function 'logLik.mkinfit' which makes it possible to calculate an AIC for mkinfit objects |