my xnew isno being displayed and how would i calculate coefficient of determination and the standard error of estimate?
1 vue (au cours des 30 derniers jours)
Afficher commentaires plus anciens
v = 10:10:80;
F = [25 70 380 550 610 830 1220 1450];
x = v(:); y = F(:);
fssr = @(a) sum((y-a(1)*sinh(a(2)*x)));
[a, ssrr] = fminsearch(fssr,[1 1])
plot(x,y,'md'); hold on; grid on;
fplot(@(x) amin(1)+x^amin(2), [0 90])
legend('data','nonlinear fit')
sr = fmin; % same as: sr = sum((y-amin(1)-x.^amin(2)).^2)or sr = fssr(amin)
RR = 1-sr/sum((y-mean(y)).^2);
syx = sqrt(sr/(length(y)-length(amin)));
xnew = 5; ynew = amin(1)+xnew^amin(2);
0 commentaires
Réponses (1)
Star Strider
le 13 Avr 2015
I can’t run your code. Your ‘amin’ variable is missing.
If you want to fit ‘y’ to sinh (shifted and scaled), I would add two parameters to get a better fit, and square the difference in your ‘fssr’ function to produce a sum-of-squares cost function:
ssinh = @(a) a(1)*sinh(a(2)*x+a(3))+a(4);
fssr = @(a) sum((y-ssinh(a)).^2);
[a, ssrr] = fminsearch(fssr,[1 1 1 1])
0 commentaires
Voir également
Catégories
En savoir plus sur Interpolation dans Help Center et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!