Generate Three (or More) Correlated Random Vectors In Matlab

% Courtesey of: User "Bruno Luong" (from MATLAB Newsgroup) % By Hammad Munawar (hammadmunawar.weebly.com),IAA,Islamabad, Pakistan % % Generate three correlated normally distributed random vectors (x,y,z) % % SD(x)=10,mean(x)=100 % SD(y)=10,mean(y)=100 % SD(z)=2.5,mean(z)=100 % corr(x,z)=0 % corr(x,y)=-0.5 % corr(y,z)=-0.5 % sd_x=10;%standard deviation sd_y=10;%standard deviation sd_z=2.5;%standard deviation mean_x=100;%mean mean_y=100;%mean mean_z=100;%mean n = 100;%lenght of vectors % % Variance matrix % S = diag([sd_x sd_y sd_z].^2); S(1,2)=-0.5*sqrt(S(1,1)*S(2,2)); S(2,3)=-0.5*sqrt(S(2,2)*S(3,3)); S(3,2)=S(2,3); S(2,1)=S(1,2); % L=chol(S);%cholesky factorization M=[mean_x;mean_y;mean_z]; %means vector % U=bsxfun(@plus,M,L.'*randn(3,n));%generate correlated random numbers U=U'; % x=U(:,1);%form vector y=U(:,2);%form vector z=U(:,3);%form vector % corr(x,y)%verify correlation corr(x,z)%verify correlation corr(y,z)%verify correlation