How to fit linear plateau equation ?
4 views (last 30 days)
Show older comments
I want to fit linear equation with plateau condition.
y=a(x+n)+b
These are the data in which I Want to fit:
x=[0,80,100,120,150],
y=[1865,3608,4057,4343,4389]
I don not know how to find the plateau value of yield.
3 Comments
Accepted Answer
Bruno Luong
on 10 Dec 2018
Edited: Bruno Luong
on 10 Dec 2018
You can use this FEX file (BSFK)

The code is by default use QUADPROG from MATLAB, but can also work with QPAS project by Adrian Wills for user who has not access to optimization toolbox as showed below:
x=[0,40,80,100,120,150,170,200];
y=[1865,2855,3608,4057,4343,4389,4415,4478];
opt = struct('KnotRemoval','none','sigma',1,'qpengine','qpas');
pp = BSFK(x,y,2,2,[],opt);
% prepare graphical data
xi = linspace(min(x),max(x));
yi = ppval(pp,xi);
xb = pp.breaks(2);
yb = ppval(pp,xb);
clf
plot(x,y,'or',xi,yi,'-b');
xline(pp.breaks(2));
text(xb,yb,sprintf('(%g,%g)',xb,yb));
3 Comments
John D'Errico
on 10 Dec 2018
Note that the upper line segment is not held constant in your solution, as it seems the OP wants.
Bruno Luong
on 10 Dec 2018
I don't know exactly what he wants, I did not follow the thread closely.
I give the code if he is not happy then I hope he'll come back with further request.
If the plateau must be constant I can enfore the derivative to 0.
More Answers (0)
See Also
Categories
Find more on Linear Least Squares in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!