GCtest.m

function [Wald,pval]=GCtest(y1,y2,p)
% Testing for the null hypothesis that y2 does not Granger cause y1
% Wald is the value of the test stat
% pval is the p-value
% B'==[v,A1,...,Ap]

k1=size(y1,2);  
k2=size(y2,2);
K=k1+k2;
y=[y1,y2];
[B,SIG,~,ZZ]=olsvar(y,p);
alpha=reshape(B,K*(1+K*p),1); % alpha=vec(B)
SIGa = kron(SIG,inv(ZZ)); %covariance of beta_hat=vec(B)y
C=[];
Cmat=zeros(K,1+K*p);
for s=1:k1   % row info
    for r=1:p
        for q=1:k2   % column info
            Ctemp=zeros(1,(1+K*p)*K);
            Ctemp(1,(s-1)*(1+K*p)+(r-1)*K+1+k1+q)=1;
            Cmat=Cmat+reshape(Ctemp,1+K*p,K)';
            C=[C;Ctemp];
        end
    end
end

Wald=(C*alpha)'*inv(C*SIGa*C')*(C*alpha);
pval=1-cdf('chi2',Wald,size(C,1));

disp('H0: y2 does not Granger cause y1')
disp('')
disp(['The p-value is' num2str(pval)])
disp('')
pos=repmat('*',K,1+K*p); 
pos((Cmat==1))=['0'];
disp(['Imposed zero restrictions are of the form'])
disp(['[v,A1,...,Ap] = '])
disp(pos)