Prev: How to run multiple m-files in a single instance of matlab?
Next: Read malformed text file
From: Mohammed Khalilia on 16 Feb 2010 11:39
I am trying to plot in 3D a dataset generated with a mean and covariance as follows:
mean=[8 2] ;
sigma=[4.1 0;0 2.8];
x = mvnrnd(mean,sigma,100) ;
[X Y]=meshgrid(x(:,1),x(:,2));
Z = peaks(X,Y);
surf(X,Y,Z); %or mesh(X,Y,Z)

The 3D does not look correct, since it is a Gaussian dataset, I expect a nice bell shape, but what I get is more like a triangle. The elements in the Z matrix almost all are zeros. So I suspect that I generated the Z matrix incorrectly.

I followed the examples from MATLAB documentation:
[X,Y] = meshgrid(-2:.2:2, -2:.2:2);
Z = X .* exp(-X.^2 - Y.^2);
surf(X,Y,Z)

Thanks for any suggestions.
From: Peter Perkins on 16 Feb 2010 14:19
On 2/16/2010 11:39 AM, Mohammed Khalilia wrote:
> I am trying to plot in 3D a dataset generated with a mean and covariance
> as follows:
> mean=[8 2] ;
> sigma=[4.1 0;0 2.8];
> x = mvnrnd(mean,sigma,100) ;
> [X Y]=meshgrid(x(:,1),x(:,2));
> Z = peaks(X,Y); surf(X,Y,Z); %or mesh(X,Y,Z)
>
> The 3D does not look correct, since it is a Gaussian dataset, I expect a
> nice bell shape, but what I get is more like a triangle. The elements in
> the Z matrix almost all are zeros. So I suspect that I generated the Z
> matrix incorrectly.

If you're expecting to see something like the density of a bivariate normal distribution, then I'd have to agree. How about this:

x = mvnrnd(mean,sigma,10000);
hist3(x)

or the example on this page in the documentation:

<http://www.mathworks.com/access/helpdesk/help/toolbox/stats/brn2ivz-89.html>

Hope this helps.
 | 
Pages: 1
Prev: How to run multiple m-files in a single instance of matlab?
Next: Read malformed text file