goodness of fit data for a fit

Hi,

I am new in matlab and I would be thankful if you help me.

I want to get the goodness of fit data using fit = goodnessOfFit(x,xref,cost_func) or any other command that provides statistics of goodness of fit for the following fit:

• load sample.mat;
• rangemin = 50;
• rangemax = 149;
• xprime(1)=1;
• xprime(2)=20;
• [x]=lsqcurvefit(@our,xprime,t(rangemin:rangemax),ourdata(rangemin:rangemax));
• y=our(x,t);
• x1=x(1,1)
• x2=x(1,2)
• function y=our(x,t)
• y=((0.282*x(2))-0.24)*x(1)*exp((x(2)-0.24)*t);
• end

I am not sure what should be x and xref and cost_func here and where I should put this command?

Answer 1

[Torsten]

Statistical information about the goodness of fit can be obtained by using "fit" instead of "lsqcurvefit" and the second output argument structure "gof":

https://de.mathworks.com/help/curvefit/fit.html

I think the function "goodnessOfFit" is insufficient for this purpose.

Answer 2

[Sulaymon Eshkabilov]

From your posted link, you can see that:

(1) x is:

x = lisqcurvefit(...) gives you the found fit model coefficients. E.g.: from your exercise:

• y=((0.282*x(2))-0.24)*x(1)*exp((x(2)-0.24)*t);

x = [x(1), (x2)]

(2) xref is:

xref is your measured data

(3) cost_func is:

cost_func = 'NRMSE' % normalized root mean squared error (NRMSE) as the cost function

You may also consider setting it to be:

cost_func = 'MSE'; % Mean squared error

Or

cost_func = 'NMSE'; % Normalized mean squared error