# How to fit this curve?

32 views (last 30 days)
Gabriel Costa on 28 Feb 2021
Commented: Star Strider on 28 Feb 2021
x = [39.224 49.03 58.836 68.642 78.448 88.254 98.06 107.866 117.672 127.478 137.284];
y = [2.4218 3.9931 6.0817 8.7791 12.1994 16.4867 21.8260 28.4584 36.7052 47.0032 59.9637];
Tatti Singh on 28 Feb 2021
Thanks for the help. I was looking the same.

Star Strider on 28 Feb 2021
The best model is the mathematical expression of the process that created those data.
A power law fit appears to provide a reasonable approximation:
x = [39.224 49.03 58.836 68.642 78.448 88.254 98.06 107.866 117.672 127.478 137.284];
y = [2.4218 3.9931 6.0817 8.7791 12.1994 16.4867 21.8260 28.4584 36.7052 47.0032 59.9637];
fcn = @(b,x) b(1).*x.^b(2);
B = fminsearch(@(b) norm(y - fcn(b,x)), rand(2,1)); % Power Law
figure
plot(x, y, '.')
hold on
plot(x, fcn(B,x), '-r')
hold off
grid
xlabel('X')
ylabel('Y')
producing:
.
##### 2 CommentsShowHide 1 older comment
Star Strider on 28 Feb 2021
As always, my pleasure!

Tatti Singh on 28 Feb 2021
Thanks for it.