Creates conditional gaussian process simulations.
Sim=SimCondGaussian (Xnew,model,ns);
data points for simulations (mx*n). Row vectors representing data points.
kriging model, mlist of type kmodel created previously.
number of simulations. Default value ns=1.
matrix containing simulated function values at data points Xnew (m*ns).
Creates Conditional Gaussian process simulations at Xnew data points, given observations and covariance model by kriging model. Note- This function uses gobal variable S_kriging, T_kriging.
// 1D example x_bound=[-2; 2]; X=(linspace(x_bound(1),x_bound(2),4))'; // Points of experiment YExp=sin(X);// Response values noise=0.0000001;theta=1; [kmodel] =km(X,YExp,0,'gaussian',theta,noise);// creates kriging model (constant trend) Xtest=(linspace(x_bound(1),x_bound(2),100))'; // 100 point test sample [Sim1]=SimCondGaussian(Xtest,kmodel,50); scf; plot(Xtest,Sim1,'k'); // simulations | ![]() | ![]() |