How to fit an integral with more than one variables?

Hello,

I'm trying to fit data with an integral with 4 variables and I'm not sure how to write the fit.

I have two matrix I(y-axis) and tau(x-axis), and here is the part of my script which fails:

syms d mu sig x
s = fitoptions('Method','NonlinearLeastSquares','Lower',[0.05 0.05],'Upper',[100 100],'Startpoint',[0.05 0.05]);
fff=@(mu,sig,x)vpaintegral((1./(d.*sig.*sqrt(6.28))).*exp((-(log(d)-mu).^2)./(2.*sig.^2)).*0.5.*(1-exp(-(d.*x).^1.1)).*cos(1.9.*d.*x),d,0.05,10)
fitint= fittype(fff, 'options',s)  
f3=fittype(tau,Indq,fitint)    %ligne 45

I get the following errors:

Error using fittype>iCreateFittype

First input argument must be a string scalar or character vector, cell array or anonymous function.

Error in fittype (line 330)

obj = iCreateFittype( obj, varargin{:} );

Error in Analyse_DQ (line 45)

f3=fittype(tau,I,fitint)

Any suggestions?

Thanks

Answer 1

[Pooja Kumari]

Hi LR,

I understand that you are facing "Error using fittype>iCreateFittype" error when you're trying to fit your data using a custom function that involves an integral.

The error you're encountering is due to incorrect usage of the "fittype" function.

f3=fittype(tau,Indq,fitint)   %line 45

Perform the fitting instead of "fittype" function

[fittedCurve, gof] = fit(tau, I, fitint);

"First input argument must be a string scalar or character vector, cell array or anonymous function." is occuring due to matrix tau(x-axis).

You can refer to the below documentation for more information on "fittype" function:

https://www.mathworks.com/help/curvefit/fittype.html#btpaend-1-linearModelTerms