Vector auto regressive (VAR) model with 2 endogeneous variables to V(AR) model with one endogenous variable

Hi Guys,

I'm trying to apply my data to a study and I use matlab to resolve my problem. The thing is that I want to transfer the original VAR model with two endogeneous variables to a VAR or AR model with only one endogeneous variable.

My code looks like this:

function [R_e, dp] = varmom(M,T,r_f);

epsilon = zeros(M,T,2);
r = zeros(M,T);

for m=1:M

      epsilon(m,:) = mvnrnd([0;0], [0.006,-0.0051;-0.0051,0.0049],T);
  end

for m=1:M

      dp0=-0.155/(1-0.958);
      r(m,1) =  0.737767+0.060*dp0 + epsilon(m,1,1);
      dp(m,1)=-0.155+0.958*dp0 + epsilon(m,1,2);

      for i = 1:T-1
          r(m,i+1) = 0.227+0.060*dp(m,i) + epsilon(m,i+1,1);
          dp(m,i+1) = -0.155+0.958*dp(m,i) + epsilon(m,i+1,2);
      end
  end

%create excess return out of log excess return
R_e = r_f*exp(r)-r_f;
disp('Excess return created out of log excess return')

But I want to go back to a VAR model which only contains this output the UMD factor.

How to reduce my code and what variables to change if I only use UMD?

Thanks in advance

Kevin