Page 1 of 1

questions about the moment calculation of news shock

PostPosted: Mon Jan 02, 2017 3:28 pm
by zhanshuo
Dear all ,
I am now working on news shock topic and are learning by studying the typical example: https://github.com/JohannesPfeifer/DSGE ... _model.mod
I am confused that, If I used the command:
simult_
that split the whole simulation into two , or to say , like starting a new simulation.
In this way we can simulate the path of an unrealized shock.
So my question is , now what the second moment in the oo file refer to? the whole simulation? the second moment after "simult_"?
and if I want to do some welfare analysis under unrealized shock circumstance , what is the right way?
Thanks.

Re: questions about the moment calculation of news shock

PostPosted: Tue Jan 03, 2017 2:37 pm
by jpfeifer
What do you exactly mean? The
Code: Select all
simult_

function allows you to simulate your model with any given sequence of shocks you choose. The second moments, in contrast, are based on the covariance matrix of shocks you specify in the
Code: Select all
shocks

block. They are the theoretical moments you get asymptotically when simulating the model with random shocks drawn from this distribution.
For welfare, what you need depends on the welfare criterion you use.