Dynare forums

[code]var y,cp,i,k,n,d,rd,p,w,ce,rb,b,l,rl,a,v,r,xn,f,ib;
varexo ea,eu;
parameters betap,shp,psi,betae,she,alpha,sigma,cd,cl,cb,oy,op,or,rhoa,rhov,ps,rds,rls,rbs,rs,ws,ys,ls,ds,bs,ks,ns,cps,ces,is,fs,ibs,xns;
betap=0.9985;
shp=1;
psi=1;
betae=0.9925;
she=1;
alpha=0.5;
sigma=0.03;
cd=0.01;
cl=0.01;
cb=0.01;
or=0.5;
op=1.5;
oy=0.125;
rhoa=0.95;
rhov=0.95;
sa=0.05;
su=0.05;
ps=1.0245;
rds=ps/betap;
rls=ps/betae;
rbs=ps/betae;
rs=(0.9*cl*cb*rds+cd*cb*rls+cd*cl*rbs)/(cl*cb*0.9+cd*cl+cd*cb);
ws=(0.5*(betae/(1-betae*(1-sigma)))^0.5)^2;
ys=((2*ws^2)/(0.5-(1-rbs/ps)*(rbs-rs)*100-(1-rls/ps)*(rls-rs)*100))^0.5;
ls=(rls-rs)*100*ys;
bs=(rbs-rs)*100*ys;
ds=0.1*(rs-rds)*100*ys;
ks=(alpha*betae/(1-(1-sigma)*betae))*ys;
ns=0.5*ys/ws;
cps=(ws^2)/(ys*0.5);
is=ks*sigma;
ces=0.5*ys+(1-rbs/ps)*(rbs-rs)*100*ys+(1-rls/ps)*(rls-rs)*100*ys-is;
fs=(rls-rs)*ls+(rbs-rs)*bs+10*(rs-rds)*ds;
ibs=0;
xns=0.1;
model;
1=exp(rd)*betap*((exp(cp(+1)-cp))^(-shp))*(1/exp(p(+1)))*(exp(v-v(-1)));
exp(w)=((exp(cp))^shp)*(exp(n))^psi;
exp(cp)+exp(d)=exp(rd(-1)+d(-1)-p)+exp(w+n)+exp(f)+0.0044778844741975;//h
y=a+k(-1)*alpha+(1-alpha)*n;
exp(k)=(1-sigma)*exp(k(-1))+exp(i);
1=exp(rb)*betae*((exp(ce(+1)-ce))^(-she))*(1/exp(p(+1)))*(exp(v(+1)-v));
1=exp(rl)*betae*((exp(ce(+1)-ce))^(-she))*(1/exp(p(+1)))*(exp(v(+1)-v));
exp(w)=(1-alpha)*(exp(y-n));
1=betae*((exp(ce(+1)-ce))^(-she))*(exp(v(+1)-v))*((alpha*exp(y-k(-1)))+1-sigma);
exp(ce)+exp(w+n)+exp(i)+exp(rb(-1)+b(-1)-p)+exp(rl(-1)+l(-1)-p)=exp(y)+exp(b)+exp(l);
exp(p)-ps=betap*(exp(p(+1))-ps);//e
exp(rl)=exp(r)+(cl/exp(y))*exp(l);
exp(rb)=exp(r)+(cb/exp(y))*exp(b);
exp(rd)=exp(r)-exp(xn)*(cd/exp(y))*exp(d)/((exp(xn))^2);
exp(f)=(exp(rl)-exp(r))*exp(l)+(exp(rb)-exp(r))*exp(b)+((1/exp(xn))*(exp(r)-exp(rd)))*exp(d)-(1/(2*exp(y)))*(cd*((exp(d-xn))^2-(ds/xns)^2)+cl*((exp(l))^2-ls^2)+cb*((exp(b))^2-bs^2));
r=or*r(-1)+(1-or)*(log(rs)+op*(p(+1)-log(ps))+oy*(y-log(ys)));
ib+exp(d-xn)=exp(l)+exp(b)+exp(d);
a=rhoa*a(-1)+ea;
v=rhov*v(-1)+eu;
exp(y)=exp(cp)+exp(ce)+exp(i);

end;
initval;
ea=0;
eu=0;
a=0;
v=0;
y=log(ys);
cp=log(cps);
i=log(is);
k=log(ks);
n=log(ns);
d=log(ds);
rd=log(rds);
p=log(ps);
w=log(ws);
ce=log(ces);
rb=log(rbs);
b=log(bs);
rl=log(rls);
l=log(ls);
r=log(rs);
f=log(fs);
ib=ibs;
xn=log(0.1);

end;
resid(4);

check;

shocks;
var ea; stderr sa;
var eu; stderr su;
end;
stoch_simul y,i,k;/code]


Residuals of the static equations:

Equation number 1 : 0
Equation number 2 : 0
Equation number 3 : -0.0008406
Equation number 4 : 0
Equation number 5 : 0
Equation number 6 : 0
Equation number 7 : 0
Equation number 8 : 0
Equation number 9 : 0
Equation number 10 : 0
Equation number 11 : 0
Equation number 12 : 0
Equation number 13 : 0
Equation number 14 : 0
Equation number 15 : 0
Equation number 16 : 0
Equation number 17 : 0
Equation number 18 : 0
Equation number 19 : 0
Equation number 20 : 0



EIGENVALUES:
Modulus Real Imaginary

9.277e-19 -9.277e-19 0
3.679e-17 -3.679e-17 0
3.041e-15 -3.041e-15 0
1.169e-13 1.169e-13 0
0.7862 0.7862 0
0.8167 -0.8167 0
0.95 0.95 0
0.95 0.95 0
0.9944 0.9944 0
1.002 1.002 0
1.002 1.002 0
1.008 1.008 0
Inf Inf 0
Inf Inf 0


There are 5 eigenvalue(s) larger than 1 in modulus
for 4 forward-looking variable(s)

The rank conditions ISN'T verified!

Error using print_info (line 39)
Blanchard Kahn conditions are not satisfied: no stable equilibrium
How to solve this ? i am a newer.

Check your timing. In particular, if

Code: Select all

1=betae*((exp(ce(+1)-ce))^(-she))*(exp(v(+1)-v))*((alpha*exp(y-k(-1)))+1-sigma);

is an Euler equation, the return to capital might be forward-looking.

jpfeifer wrote:Check your timing. In particular, if

Code: Select all

1=betae*((exp(ce(+1)-ce))^(-she))*(exp(v(+1)-v))*((alpha*exp(y-k(-1)))+1-sigma);

is an Euler equation, the return to capital might be forward-looking.

Thanks very much! I have a try ,but it doesn't work . would you like give me more help ?

As I said: check your timing. Only you know your model, so you are the only one who can do it. You must help yourself.

jpfeifer wrote:As I said: check your timing. Only you know your model, so you are the only one who can do it. You must help yourself.

Thanks! but what principle shoud i follow when checking timing？

yhysq wrote:

jpfeifer wrote:As I said: check your timing. Only you know your model, so you are the only one who can do it. You must help yourself.

Thanks! but what principle shoud i follow when checking timing？

I have tried again and again ，but it always said Blanchard Kahn conditions are not satisfied

You have to correctly enter the variables with their respective timing as is standard in economics. Predetermined variables in Dynare must be entered using the end of period stock notation. Economic theory will precisely tell you what the correct timing for each variable is.