Page 1 of 1

Choosing initial values of variables

PostPosted: Tue Dec 10, 2013 4:35 am
by ericman967
I'm new to Dynare. How do you choose initial values which are close to steady state values in equilibrium? I've included my model below. I'm only going off of other models I've seen using dynare, c=0.7, and so on. The output says the residual for equation 5 is infinity - does this mean equation 5 is the key?

Code: Select all
// The preamble
var c, Y, G, p, m;
varexo Gj Mj M;

parameters cbeta, cm;

cbeta=0.99;
cm=0.7;

// The model
model;

// Intertemporal Euler
(1/c)=cbeta * (1/(1+(p/p(-1))))*(1/c(+1))*(cm*G^(cm-1) * Gj^(cm))/(G^(cm)+Gj^(cm))^2;

// Household constraint
c*p + M + G = Y + (G^(cm)/(G^(cm)+Gj^(cm)))*(M(-1)+Mj(-1));

// MRS consumption and money
(M/p)=(c * cm*G^(cm-1)*Gj^(cm))/(cm*G^(cm-1)*Gj^(cm) - G^(cm)*(G^(cm) + Gj^(cm)));

// Money equation
(M/p)=m;

// Consumption and money condition
(cbeta/p)*(1/m) + (1/c(+1))*(cbeta/p(+1))*(G^(cm))/(G^(cm)+Gj^(cm)) - (1/c)*(cbeta/p);

end;

initval;

c= 0.7;
M= 1;
Y= 0.7;
G= 0.5;
p= 0.9;
Gj= 0.5;
Mj= 0.5;

end;
steady;


shocks;
var Gj;
periods 1:9;
values 0.1;
end;

simul(periods = 10);

Re: Choosing initial values of variables

PostPosted: Tue Dec 10, 2013 10:11 am
by jpfeifer
That's where the problem shows up. You problem starts with not initializing all variables. You only have M, but not m.