nonlinear curve fitting: how to add an extra data vector?

1 Ansicht (letzte 30 Tage)
Ramirez
Ramirez am 24 Sep. 2014
Kommentiert: Matt J am 25 Sep. 2014
Hi, I know how to fit a curve to the ' ydata ' vector, e.g.:
opt = fitoptions('Method','NonlinearLeastSquares',...
'Startpoint',[1,0.2]);
fun = fittype('p1*cos(x)+p2*sin(x)','options',opt);
[fitobject,gof,output] = fit(time',ydata,fun)
However, I would like to include an additional data vector [same size as time and ydata(time) ] as a non-free time varying parameter [ pdata(time) ] to the function, so that we have something along this form:
fun = fittype('pdata*(p1*cos(x)+p2*sin(x))','options',opt);
Does anyone know how to implement this?

Melden Sie sich an, um diese Frage zu beantworten.

Antworten (1)

Matt J
Matt J am 24 Sep. 2014
Bearbeitet: Matt J am 24 Sep. 2014
You can specify the model as an anonymous function and pass parameters to that in the usual way and/or you could use the 'problem' input option to fittype/fit, see
http://www.mathworks.com/help/curvefit/fittype.html#btpaend-18
  2 Kommentare
Ramirez
Ramirez am 24 Sep. 2014
I tried your suggestion using:
fun = fittype( @(p1,p2,pdata,x) pdata*(p1*cos(x)+p2*sin(x)))
[fitobject,gof,output] = fit(time',y,fun,'problem','pdata')
where pdata is a vector with the same length as y and time. However I get the following error message:
Error using fit>iAssertNumProblemParameters (line 1115) Wrong number of values for problem parameters. Specify the values as a cell array with one element for each problem parameter in the fittype.
Matt J
Matt J am 25 Sep. 2014
fun = fittype( @(p1,p2,x) pdata.*(p1.*cos(x)+p2.*sin(x)),'problem','pdata');
[fitobject,gof,output] = fit(time',y,fun,'problem',pdata)

Melden Sie sich an, um zu kommentieren.

Kategorien

Mehr zu Linear and Nonlinear Regression finden Sie in Help Center und File Exchange

Tags

Produkte

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by