Weighted Least Squares Including Correlation in Error
Code for performing the weighted Least Squares Including Correlation in Error (WLS-ICE) algorithm when fitting a function to ensemble averages, implemented in languages listed below, with example data with M=100 fractional Brownian motion trajectories (generated with parameters as in paper "Fitting a function to time-dependent ensemble averaged data") (or arXiv):
Python scripts that read in the data and apply the WLS-ICE algorithm to it. Run ./example.py ../data from python-folder, for a demonstration.
Octave/Matlab - code ported from Python.
d2f.mdefines the analytical function to fit, its gradient and hessian, respectively. Run
example.mfile for example.
Hy Lisp, analogous to Python.
We also provide the scripts (Python) used to generate trajectories for the four example systems.
All code is made available under GNU General Public License v3. https://www.gnu.org/licenses/gpl-3.0.html