initval and histval

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.

initval and histval

Postby walshie00 » Tue Dec 12, 2006 3:26 pm

We have some confusion regarding the use of the stoch_simul command to simulate a model with multiple lags.

Specifically, when specifying the initial state (which includes lags) when should we use histval and when should we use initval? How much information do we have to provide? Should we provide historical values for all variables or just for variables that have lags in the state. How does one determine which lags appear in the state? Does anything change if we wish to supply values for the exogenous shocks?

We have tried experimenting with a simple model but cannot get the results we expect from the simulation.

Basically, do you have any advice on using histval and initval?

- Kieran and Krishna
walshie00
 
Posts: 1
Joined: Fri Sep 29, 2006 2:24 pm

Re: initval and histval

Postby MichelJuillard » Wed Dec 13, 2006 9:16 pm

walshie00 wrote:We have some confusion regarding the use of the stoch_simul command to simulate a model with multiple lags.

Specifically, when specifying the initial state (which includes lags) when should we use histval and when should we use initval?

inital is always necessary to provide initial guess values to find the steady state of a nonlinear model.

histval is used to provide initial values for simulations starting away from the steady state. It is the only way to specify different starting values at different lags away from the initial period
How much information do we have to provide? Should we provide historical values for all variables or just for variables that have lags in the state.

in histval you can only provide values for the state variables away from the steady state
How does one determine which lags appear in the state?

all lagged variables belong to the state. It isn't necessarily a minimal state representation, but it is the simplest approach.
Does anything change if we wish to supply values for the exogenous shocks?

In a stochastic simulation, you can't supply value for exogenous shocks. Use rather a deterministic model

Best

Michel
MichelJuillard
 
Posts: 680
Joined: Thu Nov 18, 2004 10:51 am


Return to Dynare help

Who is online

Users browsing this forum: Google [Bot] and 10 guests