Page 1 of 1

Equation counting, SS condition

PostPosted: Mon Oct 27, 2014 12:26 pm
by Charles
Dear All,
I am new in Dynare and I have two (perhaps silly) questions:

1/ Equations counting :
Let
Eq1: P=f(Qs) the supply equation
Eq2: Q=f(Qd) the demand equation
Eq3: Qd = Qs the equilibrium equation
How do I account for these 3 equations in my main Dynare file “myfile.mod” and my Matlab file that computes the SS “myfile_steadystate.m”
I guess Eq1 and Eq2. In Dynare file “myfile.mod” and Eq1 (or Eq2) and Eq3. in the Matlab file “myfile_steadystate.m”

2/SS condition and Dynare “Steady.m”
Since “Steady.m” computes the SS by solving the model where no time leads and lags are accounted for, how does one add a NSS condition.
For example
in “myfile.mod”
Current_Account (t)= TB(t) + Interest(t)
In “myfile_steadystate.m” I hypothesize that TB = 0 which implies
Current_Account = Interest
My question is, when Dynare’s “steady.m” computes the SS by solving the model without leads and lags, it solves
Current_Account = TB + Interest
How can I make sure that it takes into account that I add the condition TB = 0 when I solve for the SS myself in “myfile_steadystate.m” to provide values in initval block?
I would appreciate if someone could answer me.
Charles

Re: Equation counting, SS condition

PostPosted: Wed Oct 29, 2014 10:36 am
by jpfeifer
1. I don't get your point. The way you write it down, you should have three equations in three unknowns (demand, supply and the market clearing price). All three equations are needed, both in the mod-file and the steady state to uniquely determine all three variables.
2. What is an NSS conditions? If your model is non-stationary due to a unit root. Then, there are infinitely many steady states and you need to provide Dynare with the one you want. If you use a steady state file or a steady_state_model block, you are providing an exact steady state that Dynare will take as given. It will not enter numerical steady state-finding using initial values.

Re: Equation counting, SS condition

PostPosted: Wed Oct 29, 2014 11:29 am
by Charles
Thank you very much for your reply.
1. Sorry for my messy post. Nonetheless you got my point and your answer is crystal clear.
2. I thought that Dynare uses a different file and/or method to double check the user-provided SS values. It is now clear to me that it always uses the SS file provided by the user and as a result automatically accounts for any user’s SS condition such as a zero Trade Balance.
Your answers will help me write and debug my files.
Kudos dear jpfeifer.

Re: Equation counting, SS condition

PostPosted: Wed Oct 29, 2014 12:01 pm
by jpfeifer
2. The user provided steady state file is checked, but if it does not solve the steady state, no additional search for correct values is conducted.

Re: Equation counting, SS condition

PostPosted: Wed Oct 29, 2014 1:31 pm
by Charles
Thank you very much!