Fit Experimental Data with MATLAB: Best Practices

Click For Summary

Discussion Overview

The discussion revolves around fitting experimental data to a theoretical model using MATLAB software. Participants explore various approaches to determine fit parameters without having the theoretical data directly available, focusing on the methods and challenges involved in the fitting process.

Discussion Character

  • Exploratory, Technical explanation, Debate/contested, Homework-related

Main Points Raised

  • One participant inquires about the best way to fit experimental data using MATLAB without having the theoretical data.
  • Another suggests plotting experimental data against theoretical data to identify discrepancies.
  • A participant mentions they lack theoretical data but have a model and are seeking fit parameters.
  • It is proposed that running the model with input parameters from the experiment and plotting the output could be a solution.
  • One participant notes that they used least squares curve fitting but found that the fit parameters vary as a function of the input.
  • A later reply emphasizes the need for a clear description of the problem, including details about the theoretical model and data, suggesting that fitting methods depend heavily on context.

Areas of Agreement / Disagreement

Participants express varying opinions on how to approach the fitting problem, and there is no consensus on a single best method. The discussion remains unresolved regarding the specifics of the theoretical model and the appropriate fitting techniques.

Contextual Notes

Participants highlight the importance of context in selecting fitting methods, indicating that a lack of detailed information may hinder effective solutions. The discussion also reflects uncertainty about the nature of the theoretical model and the data involved.

sounouhid
Messages
20
Reaction score
2
Hello everyone
help please
I need to fit experimental data with a theoretical model using the MATLAB software what is the best way to do it?
thx for advance
 
Physics news on Phys.org
Why not try plotting the experimental data vs the theoretical data to see where they differ?
 
thank you
but i don't have the the theoretical data .i just have the model ,I'm looking for the fit parameters
 
Can't you run the model with input parameters matching the ones recorded when you did the experiment and then plot the relevant model output?
 
yes i did it using the least squares curve fit ,but the fit parameters are not constant they are function of the input.
 
@ sounouhid : One cannot answer to your question without a clear description of your problem. What kind of "theoretical model" are you talking about ? What kind of data ? And more details on the context of the problem are essential. The fitting methods are numerous and depends on the context. A whole book should not large enough to answer if the problem is not restricted. For example, some procedures of fitting to several models are available in : https://fr.scribd.com/doc/14674814/Regressions-et-equations-integrales .(paper in French with some parts in English. All procedures are understandable in English)
 
  • Like
Likes   Reactions: sounouhid

Similar threads

  • · Replies 12 ·
Replies
12
Views
4K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 5 ·
Replies
5
Views
3K
  • · Replies 6 ·
Replies
6
Views
1K
Replies
6
Views
4K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 1 ·
Replies
1
Views
2K