- Have lsqnonlin pass in a set of parameters x.
- in your ODE solving function solve each ODE with its own initial conditions for its own set of times.
- Concatenate all of the solutions as F(x) and calculate F(x) - [xdata;ydata;zdata] as the lsqnonlin output (response)
lsqnonlin multiple dataset optimisation
3 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Malgorzata Wieteska
am 27 Feb. 2021
Kommentiert: Malgorzata Wieteska
am 2 Mär. 2021
Hello;
I would like to optimise simple ode using multiple datasets. I wonder if is possible to do it using lsqnonlin command? Of course, I will want to allow for different values of the initial conditions for each dataset.
I found, that Population-wide parameter estimation can be done in the Simbiology, however, I would rather prefer to use the lsqnonlin.
Thank you in advance
0 Kommentare
Akzeptierte Antwort
Alan Weiss
am 1 Mär. 2021
It sounds like you are trying to find a set of ODE parameters that work for multiple data sets. If that is true, then I think that the workflow is this:
Good luck,
Alan Weiss
MATLAB mathematical toolbox documentation
Weitere Antworten (2)
Arthur Goldsipe
am 27 Feb. 2021
You can use lsqnonlin with or without SimBiology. If you want to do this without SimBiology, you need to write a function that takes in the parameters you are trying to estimate, solves your ODEs (for example, using ode45 or ode15s, and returns a vector of residuals (the difference between the simulation results and the measured values). You will also find many similar questions posted on MATLAB Answers, several of which are mentioned in the answer here.
0 Kommentare
Siehe auch
Kategorien
Mehr zu Scan Parameter Ranges finden Sie in Help Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!