Additional algorithm "d_c", more tests, docs

The new algorithm tries direct optimization of the likelihood, as well
as a three step procedure. In this way, we consistently get the
model with the highest likelihood for SFO, DFOP and HS for all 12
new test datasets.

1 files changed, 9 insertions, 7 deletions

diff --git a/docs/news/index.html b/docs/news/index.html
index 11414ed3..df6d53db 100644
--- a/docs/news/index.html
+++ b/docs/news/index.html
@@ -60,7 +60,7 @@
       </button>
       <span class="navbar-brand">
         <a class="navbar-link" href="../index.html">mkin</a>
-        <span class="version label label-default" data-toggle="tooltip" data-placement="bottom" title="Released version">0.9.49.4</span>
+        <span class="version label label-default" data-toggle="tooltip" data-placement="bottom" title="Released version">0.9.49.5</span>
       </span>
     </div>
 
@@ -122,13 +122,14 @@
       
     </div>
 
-    <div id="mkin-0-9-49-4-2019-05-08" class="section level1">
+    <div id="mkin-0-9-49-5-2019-06-04" class="section level1">
 <h1 class="page-header">
-<a href="#mkin-0-9-49-4-2019-05-08" class="anchor"></a>mkin 0.9.49.4 (2019-05-08)<small> Unreleased </small>
+<a href="#mkin-0-9-49-5-2019-06-04" class="anchor"></a>mkin 0.9.49.5 (2019-06-04)<small> Unreleased </small>
 </h1>
 <ul>
-<li><p>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</p></li>
-<li><p>The argument ‘reweight.method’ to mkinfit and mmkin is now obsolete, use ‘error_model’ instead</p></li>
+<li><p>Several algorithms for 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 default algorithm tries direct minimization and a three step procedure, and returns the model with the highest likelihood.</p></li>
+<li><p>The argument ‘reweight.method’ to mkinfit and mmkin is now obsolete, use ‘error_model’ and ‘error_model_algorithm’ instead</p></li>
+<li><p>Add a test that checks if we get the best known AIC for parent only fits to 12 test datasets. Add these test datasets for this purpose.</p></li>
 <li><p>New function ‘mkinerrplot’. This function is also used for residual plots in ‘plot.mmkin’ if the argument ‘resplot = “errmod”’ is given, and in ‘plot.mkinfit’ if ‘show_errplot’ is set to TRUE.</p></li>
 <li><p>Remove dependency on FME, only use nlminb for optimisation</p></li>
 <li><p>Use the numDeriv package to calculate hessians</p></li>
@@ -136,6 +137,7 @@
 <li><p>The code for manual weighting was removed</p></li>
 <li><p>The fitting time reported in the summary now includes the calculation of the hessians</p></li>
 <li><p>Adapt tests</p></li>
+<li><p>Fix an error in the FOCUS chi2 error level calculations that occurred if parameters were specified in parms.ini that were not in the model. A warning was already issued, but when fitting via mmkin this could easily go unnoticed.</p></li>
 </ul>
 </div>
     <div id="mkin-0-9-48-1-2019-03-04" class="section level1">
@@ -696,7 +698,7 @@
     <div id="tocnav">
       <h2>Contents</h2>
       <ul class="nav nav-pills nav-stacked">
-        <li><a href="#mkin-0-9-49-4-2019-05-08">0.9.49.4</a></li>
+        <li><a href="#mkin-0-9-49-5-2019-06-04">0.9.49.5</a></li>
         <li><a href="#mkin-0-9-48-1-2019-03-04">0.9.48.1</a></li>
         <li><a href="#mkin-0-9-47-5-2018-09-14">0.9.47.5</a></li>
         <li><a href="#mkin-0-9-47-3">0.9.47.3</a></li>
@@ -740,7 +742,7 @@
 </div>
 
 <div class="pkgdown">
-  <p>Site built with <a href="https://pkgdown.r-lib.org/">pkgdown</a> 1.3.0.9000.</p>
+  <p>Site built with <a href="https://pkgdown.r-lib.org/">pkgdown</a> 1.3.0.</p>
 </div>
       </footer>
    </div>