sim1_linear.m: Fix evaluation of dynamic model at deterministic steady state

Did not correctly account for exogenous variables being potentially present with leads and lags

matlab/perfect-foresight-models/sim1_linear.m

@@ -111,7 +111,8 @@ dynamicmodel = str2func([M.fname,'_dynamic']);
 z = steadystate_y([ip; ic; in]);
 
 % Evaluate the Jacobian of the dynamic model at the deterministic steady state.
-[d1,jacobian] = dynamicmodel(z, transpose(steadystate_x), params, steadystate_y, 1);
+[d1,jacobian] = dynamicmodel(z, repmat(transpose(steadystate_x),options.periods+M.maximum_lag+M.maximum_lead,1),...
+    params, steadystate_y, M.maximum_lag+1);
 
 % Check that the dynamic model was evaluated at the steady state.
 if max(abs(d1))>1e-12