gaussian fit to the data

10 次查看（过去 30 天）
Megha
Megha 2019-4-16
评论： Torsten 2019-4-16
I have my data like
x = 1:1934;
y = (Attached below);
To fit a gaussian distribution, I used
figure; plot(x,y);
myfittype = fittype('B.*exp(-(x-mu).^2/(2*sigma^2))',...
'dependent',{'y'},'independent',{'x'},'coefficients',...
{'B','mu','sigma'}); %
[myfit2, goodness2, output2] = fit(x,y,myfittype);
hold on;
plot(myfit2,x,y);
Even though it fits very incorrectly, I can see a Warning: Start point not provided, choosing random start point.
Ideally, alongwith mu and sigma, coefficent B should also be generated automatically while running fittype. However, it does not.
Any suggestions to improve it?
  1 个评论
Torsten
Torsten 2019-4-16
I think you mix up "maximum likelihood estimates" and "curve fitting".
Take a look at
https://de.mathworks.com/help/stats/mle.html

请先登录，再进行评论。

请先登录，再回答此问题。

回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Get Started with Curve Fitting Toolbox 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by