covariance of weighted multidimensional samples

4 views (last 30 days)
PChoppala
PChoppala on 24 Jun 2016
Edited: PChoppala on 24 Jun 2016
I have a multidimensional weighted sample set, e.g.
D=2; % dimension
N=100; % number of samples
x=randn(D,N); % samples
w=ones(1,N)/N; % corresponding weights
I would like to find the weighted covariance of this sample set. To obtain this, I first computed the weighted mean using the formula \mu = \sum_{i=1}^{N} w_{i} x_{i} as
mu=sum(bsxfun(@times,w,x),2);
Then I need to find the covariance according to the formula \Sigma = \sum_{i=1}^{N} w_{i} (x_{i} - \mu)*(x_{i} - \mu)'. My current code is this:
ct=bsxfun(@minus,particles,mu);
P=zeros(Dx,Dx,N);
for n=1:N,
P(:,:,n)=w(n)*(ct(:,n)*ct(:,n)');
end
Sigma=sum(P,3);
What is the computationally best coding procedure to calculate this covariance? Suggestions appreciated, thanks.

Answers (0)

Categories

Find more on MATLAB in Help Center and File Exchange

Tags

Community Treasure Hunt

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

Start Hunting!