How to get the coefficient of the function using non-linear regression analysis?

I coded like the picture to get the coefficient of the function with data by using non-linear regression analysis. but I don't know why the errors continue. What is the problem with that??

Answer 1

[Alan Stevens]

Needs to be more like this:

xm = 0:0.5:5;
ym = [0 0.223904 0.323333 0.372308 0.359143 0.332033 0.278462 0.243394 ...
     0.232222 0.198334 0.172414];
 
a0 = [1 1 1 1 1];
a = fminsearch(@(a)fssr(a,xm,ym), a0);
disp(a)

x = linspace(0,5,100);
y = a(1)*x./(a(2) + a(3)*x + a(4)*x.^2 + a(5)*x.^3);

plot(xm,ym,'o',x,y),grid 
xlabel('x'),ylabel('y')
legend('data','curvefit')
 
 function f = fssr(a, xm, ym)
 
          yp = a(1)*xm./(a(2) + a(3)*xm + a(4)*xm.^2 + a(5)*xm.^3);       
          f = norm(ym-yp);

 end

[skdfjhoiashfiWHGOFA]

What shoul I do just to get the coefficient without plotting?

[skdfjhoiashfiWHGOFA]

sorry *should

[skdfjhoiashfiWHGOFA]

Oh I get it!

thank you

[Alan Stevens]

Just delete the plotting from the above if you don't want it. The

disp(a)

command displays the values of a in the command window.

[skdfjhoiashfiWHGOFA]

I'm so sorry to keep asking, but how to get the maximum value of that function?

[Alan Stevens]

Just use

max(y)

[skdfjhoiashfiWHGOFA]

I've already done that but there's an error...

[Alan Stevens]

What's the error? When I do it I get 0.3673 for the max value.

[skdfjhoiashfiWHGOFA]

I solved it. I get that one too. Thank you sooo much !!