Gaussian mixture distribution plot

I'm trying to bild a gaussian mixture distribution. I'm quite new on Matlab, but I think the script is ok. I'm not able to plot my distribution. Here you can find what I have. Many thanks for your answers, cheers

%quadmodal mixture
mu=[30 0 0 0; 22 0 0 0;12 0 0 0;4 0 0 0]; 
sigma=cat(3,[9^2 0 0 0;0 6.6^2 0 0;0 0 3.6^2 0;0 0 0 1.2^2],[1 0 0 0;0 1 0 0;0 0 1 0;0 0 0 1],[1 0 0 0;0 1 0 0;0 0 1 0;0 0 0 1],[1 0 0 0;0 1 0 0;0 0 1 0;0 0 0 1]);
p=ones(1,4)/4;
obj = gmdistribution(mu,sigma,p);
ezsurf(@(x,y)pdf(obj,[x y]),[-25 25],[-25 25])

[zepp]

your mixture distribution has four dimensions, so you cannot plot that using ezsurf. Are you sure you have initialised the mu and sigma values correctly? Or is it just a 1 dimensional 4 component Gaussian distribution?

Answer 1

[Prateekshya]

Hello Marco,

It looks like you're attempting to create and plot a Gaussian Mixture Model (GMM) in MATLAB using the gmdistribution function. Your script is mostly correct, but there are a few adjustments needed to properly plot the distribution.

Here is an updated version of your script with explanations:

% Define the means of the Gaussian components
mu = [30 0; 22 0; 12 0; 4 0];

% Define the covariance matrices for each component
sigma = cat(3, ...
    [9^2 0; 0 6.6^2], ...
    [1 0; 0 1], ...
    [1 0; 0 1], ...
    [1 0; 0 1]);

% Define the mixing proportions for each component
p = ones(1, 4) / 4;

% Create a Gaussian Mixture Model object
obj = gmdistribution(mu, sigma, p);

% Plot the Gaussian Mixture Model
% Define a grid of points over which to evaluate the GMM
x = linspace(-25, 35, 100);
y = linspace(-25, 35, 100);
[X, Y] = meshgrid(x, y);

% Evaluate the GMM PDF over the grid
Z = pdf(obj, [X(:) Y(:)]);
Z = reshape(Z, size(X));

% Plot the surface
figure;
surf(X, Y, Z);
xlabel('X-axis');
ylabel('Y-axis');
zlabel('Probability Density');
title('Gaussian Mixture Model');

Ensure that the covariance matrices are the correct size for the means you have specified. Each covariance matrix should be 2x2 since your means are 2-dimensional. Use meshgrid to create a grid of points over which to evaluate the GMM. This allows you to compute the probability density function (PDF) values over a range of x and y values. After evaluating the PDF, reshape the resulting vector Z back into a matrix that matches the grid dimensions for plotting. Use surf instead of ezsurf for more control over the plot. This also allows you to label the axes and add a title. The above code produces the below result:

I hope this helps!