NKPC coding in Dynare

[Vermandel]

Hello,

I'm currently working on a model with the NKPC device, but I get a bit confused with they way of coding it. I use this functional form,

Code: Select all

pi=beta*pi(+1)+(((1-theta*beta)*(1-theta))/theta)*cm;

but regarding inflation equation, should i used this one,

Code: Select all

pi_h=py_h-py_h(-1);

or this one,

Code: Select all

pi_h(+1)=py_h(+1)-py_h;

Results are strongly different, I don't get why.

[jpfeifer]

Results are different, because Dynare treats everything with a (+1) as being conditional expectations. Hence, the first equation holds at every point in time, while the second one holds in expectations.

Code: Select all

pi_h(+1)=py_h(+1)-py_h;

actually is

Code: Select all

E_t(pi_h(+1))=E_t(py_h(+1))-py_h;

[Vermandel]

Thanks for you answer jpfeifer.

In your opinion, should I use the first or second one equation ? I get better results with the second one.

[jpfeifer]

Use the one that is consistent with your model.