How to do curve fitting to a custom model with two independent variables?
8 views (last 30 days)
Show older comments
I have data in which response is dependent upon two variables. Three data series are given:[x1, y1], [x2, y2], [x3, y3]. On excel (file attached) I have made XY scatter plot which gives three response curves. I want to fit these 3 curves to a custom model in the form: lnY = k*lnX + (a/T) + b This model is used to describe such data in our field. I just don't know how I will fit all three curves to this model? Please help!
0 Comments
Answers (1)
Matt J
on 24 May 2015
Edited: Matt J
on 24 May 2015
The independent variables are X and T? If you insist on using the Curve Fitting Toolbox, the thing to do would be to regard this as a surface in X and T and use sfit().
It is noteworthy, though, that your model is linear lnX and 1/T. So, you could really probably do a simple linear fit with mldivide(). In other words, just solve the linear equations for k, a, and b using normal linear algebraic operations.
2 Comments
See Also
Categories
Find more on Fit Postprocessing in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!