How do i show the "delta" difference between the two plotted curves?

I want to analyze the difference between the two plotted curves(fig 1) from two different codes. The zoomed-in version (fig 2). Is there any way i can get the "delta" difference in a much better and clearer manner since the coordinates are very close to each other? Maybe an interp function ? My code and .mat files are provided below. Any help is appreciated!

load coords_mine.mat

load coords_sir.mat

plot(a,b ,'.g') % code 1

hold on

axis equal

plot(x2,y2,'.b') % code 2

axis equal

delta = abs((b - y2)./y2) * 100;

xCenter = 1:1:length(b);

yCenter = (b+y2)/2;

for k = 1 : length(xCenter)

textLabel = sprintf('%.1f', delta(k));

text(xCenter(k), yCenter(k), textLabel, 'HorizontalAlignment', 'center')

end

Answer 1

[Mathieu NOE]

hello

you need to "realign" your two data sets vs common x vector - in other words we need to creat a new x vector and interpolate both y datas vs this new x vector - here I opted for pchip interpolation method.

the bottom plot is your delta vector , I picked only a few values to be displayed on the top plot

hope it helps !!

load coords_mine.mat
load coords_sir.mat

% take unique only x data 
[a,ia,ic] = unique(a);
b = b(ia);

[x2,ia,ic] = unique(x2);
y2 = y2(ia);

% plot
figure(1);
subplot(211),
plot(a,b ,'.g') % code 1
hold on
plot(x2,y2,'.b') % code 2

% create common x axis data 
xmin = max(a(1),x2(1));
xmax = min(a(end),x2(end));
xnew = linspace(xmin,xmax,300)';

% and interpolate y data using pchip method
bnew = interp1(a,b,xnew,'pchip');
y2new = interp1(x2,y2,xnew,'pchip');

delta = abs((bnew - y2new)./y2new) * 100;

xCenter = 1:30:length(bnew); % NB decimation factor 30 vs 300 original data points => 10 values displayed
yCenter = (bnew+y2new)/2;
yCenter = yCenter(xCenter);

for k = 1 : numel(xCenter)
	textLabel = sprintf('%.3f', delta(xCenter(k)));
	text(xnew(xCenter(k)), yCenter(k)*1.1, textLabel, 'HorizontalAlignment', 'center')
end
plot(xnew(xCenter),yCenter,'xk','markersize',15)

subplot(212),
plot(xnew,delta,'b') % 

[Bamelari Jovani]

Thank you so much!

[Mathieu NOE]

as always, my pleasure !