Some questions regarding simulation

This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location where you will have to reset your password.
Forum rules
This forum is closed. You can read the posts but cannot write. We have migrated the forum to a new location (https://forum.dynare.org) where you will have to reset your password.

Some questions regarding simulation

Postby amuller » Mon Dec 28, 2015 11:59 am

I intend to do an experiment from a calibrated model that consists in simulating how inflation and inflation expectations evolve during 80 periods. I have a vector with a given shock in each of these periods.

I tried to use the EXPECTATION feature in a deterministic model to recover inflation expectations, but the output is just the lagged inflation. As far as I undestood Dynare manual, this is because the agent is supposed to know all the future path of the shocks. Is it possible to recover expectations supposing that the agent does not know the subsequent path of the shocks?

I was wondering if a stochastic simulation is the correct way to do this kind of simulation, but the agent would still know the future shocks trajectory (in this case, zero). Maybe it would work if I am able to inform the shocks vector in a stochastic context, but I also couldn´t find a way to do that.

Thanks in advance,
amuller
 
Posts: 8
Joined: Sat Oct 31, 2015 1:02 am

Re: Some questions regarding simulation

Postby jpfeifer » Wed Dec 30, 2015 10:51 am

You need stochastic simulations, because only there expectations formation is non-trivial. Then, you have to ask yourself whether the shock series for your model is perfectly anticipated.
------------
Johannes Pfeifer
University of Cologne
https://sites.google.com/site/pfeiferecon/
jpfeifer
 
Posts: 6940
Joined: Sun Feb 21, 2010 4:02 pm
Location: Cologne, Germany

Re: Some questions regarding simulation

Postby amuller » Wed Dec 30, 2015 1:06 pm

Hello mr. Pfeifer.

Thank you very much for your last answer, although I am still in trouble to solve my problem. Below, I present the whole problem:

I want to treat the model as if it was the true data generating process (DGP) of economic variables, and assess what an empirical researcher, given outcomes from this DGP, would conclude about the behaviour of inflation expectations. To this end, Monte Carlo simulations will be performed 1000 times. In each of these simulations, a random realization of 80 periods is generated for two different and uncorrelated shocks to the interest rate rule. Then, the model is solved according to this shocks information and inflation rate and its expectations should be stored for each period.

To my mind, it means that economic agents does not perfectly anticipate the shocks.

Could you please provide some information on what dynare features should I use to implement this?

thank you again.
amuller
 
Posts: 8
Joined: Sat Oct 31, 2015 1:02 am

Re: Some questions regarding simulation

Postby jpfeifer » Wed Dec 30, 2015 3:39 pm

------------
Johannes Pfeifer
University of Cologne
https://sites.google.com/site/pfeiferecon/
jpfeifer
 
Posts: 6940
Joined: Sun Feb 21, 2010 4:02 pm
Location: Cologne, Germany

Re: Some questions regarding simulation

Postby amuller » Wed Jan 06, 2016 9:46 pm

Thank you very much, it helped a lot.
amuller
 
Posts: 8
Joined: Sat Oct 31, 2015 1:02 am


Return to Dynare help

Who is online

Users browsing this forum: Google [Bot] and 12 guests