Plotting Two Gaussians on One Histogram

Question

Michael 2013-9-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/87443-plotting-two-gaussians-on-one-histogram

I'd like to be able to make a histogram which has two distinct gaussian curves.. like a camel with two humps on its back. Preferably creating the gaussians from randn() would be ideal.

I think the question is based on generating the gaussian and adding the index/bins to the vector/data in some way. So, if I've phrased this correctly: How do I tack on the indices to the gaussians and overlap them (and I mean like literal addition by that)?

The picture quality is kinda bad but this:

http://www.csse.uwa.edu.au/~pk/research/matlabfns/PhaseCongruency/Docs/GaborDCproblem.png

is what I'm going for. Except I'd like it to be binned like a histogram.. perhaps using bar would be more effective?? My knowledge is low in this area and I appreciate all the help you can offer!

Thanks in advance, M.A.B.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Laurent 2013-9-17

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/87443-plotting-two-gaussians-on-one-histogram#answer_97000

在 MATLAB Online 中打开

Hi,

you can create the two shifted histograms with randn.

gauss1=randn(100000,1)/2+1; %create gaussian around +1
gauss2=randn(100000,1)/2-1; %create gaussian around -1
gausssum=[gauss1;gauss2]; %combine values

The histogram can than be made with 'hist'

[N,X]=hist(gaussum,100);
plot(X,N);

Is this what you were looking for?

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Michael 2013-9-17

Perfect! Thank you. I was trying something like this on saturday... I should review it to see where I went wrong. Thank you very much for your input!

请先登录，再进行评论。

Answer 2

dpb 2013-9-17

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/87443-plotting-two-gaussians-on-one-histogram#answer_97003

在 MATLAB Online 中打开

Basic idea; I've got to run so you can fill in the inbetween by adding the appropriate bin counts and drawing the line on top or however you want to do it...

a=randn(200,1);   % generate a normal sample mean 0, var 1
[n1,x1]=hist(1+a*0.25); [n2,x2]=hist(0.25*a-1);  
bar(x1, n1,0.4), hold on, bar(x2,n2,0.4), hold off

Adjust the multiplier on the variance as per your data...