Combine two or more gaussian components into one Gaussian Mixture (gmdistribution)

21 views (last 30 days)
Hello,
Is there any way to combine two (or more) separate Gaussian Components into one?
clear; close all; clc;
N = 25; % N-points
D = 2; % D-dimensions
% generate data
X = randn(N,D);
Y = randn(N,D)+2; % +2 to spatially separate data components
% model gaussian components
muX = mean(X,1)
SX = cov(X)
muY = mean(Y,1)
SY = cov(Y)
GX = gmdistribution(muX,SX,1) % first component
GY = gmdistribution(muY,SY,1) % second component
% COMBINE !
GC = ???????????????(GX,GY);
I would like to get output like this:
GC =
Gaussian mixture distribution with 2 components in 2 dimensions
Component 1:
Mixing proportion: 0.500000
Mean: -0.0017 -0.1167
Component 2:
Mixing proportion: 0.500000
Mean: -0.0037 0.1367

Accepted Answer

Adam
Adam on 10 Feb 2020
Edited: Adam on 10 Feb 2020
Doesn't
GC = gmdistribution( [muX; muY], cat( 3, SX, SY ), [1; 1] );
work?

More Answers (0)

Categories

Find more on Statistics and Machine Learning Toolbox in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!