Ignore the "if true" and "end" part, that is not a part of the code. Also TF1000 was to check for outliers - So no worries there.
Barplot with a lognormal distribution
2 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
Mads Søndergaard Larsen
am 15 Mai 2018
Kommentiert: Mads Søndergaard Larsen
am 17 Mai 2018
Hi, I have this barplot of some particles measured, fitted with a lognormal distribution. However, the y-values are too large. I have tried to normalize the data with trapz(n1000,x1000), but it did not work. Is it something to do with the bins? I have tried to vary this accordingly, for nbins=20, y did not vary as much.
[n1000,x1000]=hist(vol1000,);
n1000=n1000/length(vol1000)/diff(x1000(1:2));
bar1000=bar(x1000,n1000,'hist');
hold on
TF1000=isoutlier(vol1000,'mean');
mu1000=mean(log(vol1000));
std1000=std(log(vol1000));
ix1000=linspace(min(vol1000),max(vol1000));
iy1000=pdf('lognormal',ix1000,mu1000,std1000);
figure(7)
plot(ix1000,iy1000,'r')
xlabel('Volumes [µm^3]')
legend('Particle sizes','Log normal distribution')
title('Lognormal distribution of 1000 µL Ag seed prisms')
grid on
hold off
I hope someone can help. Thank you! Regards, Mads
3 Kommentare
dpb
am 15 Mai 2018
Bearbeitet: dpb
am 15 Mai 2018
It's a sorry interface for code; granted, but to enter code, highlight the lines wanted and the press the '{}Code' icon...or two spaces before first line of paragraph has same effect; lose the effect with blank line that way, though.
As far as the actual Q?; not much we can do w/o the data or at least the figure to see what of which you speak...
Antworten (0)
Siehe auch
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!