| 1 |
var y, c, k, a, h, b; |
| 2 |
varexo e, u; |
| 3 |
|
| 4 |
parameters beta, rho, alpha, delta, theta, psi, tau; |
| 5 |
|
| 6 |
alpha = 0.36; |
| 7 |
rho = 0.95; |
| 8 |
tau = 0.025; |
| 9 |
beta = 0.99; |
| 10 |
delta = 0.025; |
| 11 |
psi = 0; |
| 12 |
theta = 2.95; |
| 13 |
phi = 0.1; |
| 14 |
|
| 15 |
model(use_dll); |
| 16 |
c*theta*h^(1+psi)=(1-alpha)*y; |
| 17 |
k = beta*(((exp(b)*c)/(exp(b(+1))*c(+1))) |
| 18 |
*(exp(b(+3))*alpha*y(+2)+(1-delta)*k)); |
| 19 |
y = exp(a)*(k(-1)^alpha)*(h^(1-alpha)); |
| 20 |
k = exp(b)*(y-c)+(1-delta)*k(-1); |
| 21 |
a = rho*a(-1)+tau*b(-1) + e; |
| 22 |
b = tau*a(-1)+rho*b(-1) + u; |
| 23 |
end; |
| 24 |
|
| 25 |
markov_switching(chain=1, number_of_regimes=2, duration=2.5, parameters=[alpha, delta, theta], number_of_lags=1); |
| 26 |
alpha.prior (shape=gamma, mean=3.22, variance=0.1^2); |
| 27 |
rho.prior (shape=uniform, mean=322, variance=2^.33, domain=[0.36, 0.88]); |
| 28 |
std(e).prior (shape=beta, mean=0.3, variance=0.1^2, domain=[-0.1 006]); |
| 29 |
std(y).prior (shape=beta, mean=0.3, variance=0.1^2, domain=[01 4]); |
| 30 |
std(c).prior (shape=beta, mean=0.3, variance=0.1^2); |
| 31 |
corr(y,c).prior(shape=beta, mean=0.3, variance=0.1^2, mode=33); |
| 32 |
corr(b,c).prior(shape=beta, mean=0.3, variance=0.1^2); |
| 33 |
corr(e,u).prior(shape=beta, mean=0.3, variance=0.1^2); |
| 34 |
alpha.options(init=1); |
| 35 |
rho.options(init=1); |
| 36 |
beta.options(init=0.2); |
| 37 |
std(u).options(init=3); |
| 38 |
corr(y,c).options(init=.02); |
| 39 |
|
| 40 |
[alpha , beta , rho].prior(shape=beta, mean=[2 3 4], variance=[[1 2 3],[2 3 4]]); |