Effacer les filtres
Effacer les filtres

Plot polyfit R-squared

31 vues (au cours des 30 derniers jours)
gorilla3
gorilla3 le 23 Août 2019
Commenté : dpb le 23 Août 2019
Hi guys,
could you please help me figure out why the polyfit works in one figure but not in the other? I don'T understand why it's not plotting the full line in the second figure.
Thank you!
load workspace_23_08_19
coeffs = polyfit(d_rel_mM, dil_rel, 1);
fittedX = linspace(min(d_rel_mM), max(d_rel_mM), 200);
fittedY = polyval(coeffs, fittedX);
coeffs_a = polyfit(d_rel_mM, time_a, 1);
fittedXA = linspace(min(time_a), max(time_a), 200);
fittedYA = polyval(coeffs_a, fittedXA);
figure(1)
plot(d_rel_mM,dil_rel,'x','Markersize',15, 'Linewidth',6)
xlabel('Imaging depth [\mum]', 'Fontname','LM Roman 10','Fontsize', 40)
ylabel('Peak dilation [%]','Fontname','LM Roman 10','Fontsize', 40)
set(gca,'Fontname','LM Roman 10','FontSize',20,'box','off')
axis tight
hold on;
plot(fittedX, fittedY, 'k-', 'LineWidth', 2);
%%
figure(3) %% INCOMPLETE LINE, SEE ATTACHED
plot(d_rel_mM,time_a,'x','Markersize',15, 'Linewidth',6)
xlabel('Imaging depth [\mum]', 'Fontname','LM Roman 10','Fontsize', 40)
ylabel('Time constant 1/\alpha [s]','Fontname','LM Roman 10','Fontsize', 40)
set(gca,'Fontname','LM Roman 10','FontSize',20,'box','off')
axis tight
hold on;
plot(fittedXA, fittedYA, 'k-', 'LineWidth', 3);

Réponse acceptée

dpb
dpb le 23 Août 2019
Because
>> fittedXA(1),fittedXA(end)
ans =
10
ans =
20
>>
and that's what you plotted as the abscissa for some reason whereas the data are plotted versus ImagingDepth which has range of
>> min(d_rel_mM),max(d_rel_mM)
ans =
0
ans =
345.4000
>>
You've mixed up x and y for the prediction line evaluation...use
coeffs_a = polyfit(d_rel_mM, time_a, 1);
fittedXA = linspace(min(d_rel_mM), max(d_rel_mM), 200);
fittedYA = polyval(coeffs_a, fittedXA);
and all will be well...
  2 commentaires
gorilla3
gorilla3 le 23 Août 2019
Ahhh wonderful! Thank you :)
dpb
dpb le 23 Août 2019
Of course, you DO realize that since is a first order fit, you only need two points to draw the line, right? :)

Connectez-vous pour commenter.

Plus de réponses (0)

Catégories

En savoir plus sur Linear and Nonlinear Regression 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!

Translated by