Can anyone suggest something that could fix the program

clear all
 clc
 %following line calls my data
 [xdataini, ydataini] = textread('data.txt','%f %f','headerlines',0);
 figure;
 plot(xdataini,ydataini,'ok')
   
 ylim([0,1]);
 xlim([0,0.25]);
 %warning off MATLAB:divideByZero
 xdata=xdataini(1:245);
 ydata=ydataini(1:245);
 x0=[1,1];
 [x]=lsqcurvefit(@subprogramTL,x0,xdata,ydata);
 V=subprogramTL([x],xdata);
 fid = fopen('data1.txt','w');
 fprintf(fid,'%3.5f \n',V);
 fclose(fid);
 hold on
 plot (xdata,V,'r')
 hold off

SUBPROGRAM

function F=subprogramTL(x,xdata)
m = 5.8;
v = 1.7;
F=(1-((x(1)/2).*atan((2.*m.*v)/((1+2.*m)^2 + v^2)*(x(2)./(2*xdata)) + 1 + 2.*m + v.^2))).^2; 
%+ ((x(1)/4).*log(((1+((2*m)./(1+(2*xdata)/x(2)))).^2 + v.^2)/((1+(2*m)).^2 +v.^2)).^2);
end

[Alex Mcaulley]

What is your issue? Are you getting any error?

[Titu Thomas]

There is no error while running the code, but the curve does not fit with the data points

Answer 1

[Prabhan Purwar]

Code is working perfectly fine. The output of lsqcurvefit() function is the best fit of variables and depends upon assigned model. For more information refer to following link:

https://www.mathworks.com/help/optim/ug/lsqcurvefit.html

As the data appears to be like an exponential function, then I would suggest you make use of fit() function with appropriate fittype and fitoption as explained in the following example.

fo = fitoptions('Method','NonlinearLeastSquares',... 
    'Lower',[0,0],... 
    'Upper',[Inf,max(cdate)],... 
    'StartPoint',[1 1]); 
ft = fittype('a*(x-b)^n','problem','n','options',fo); 
[curve2,gof2] = fit(cdate,pop,ft,'problem',2) 

For more information refer to following link:

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