Page 1 of 1

Phantom shock for IRF purpose

PostPosted: Sun Oct 19, 2014 8:32 pm
by hewei2004
Hi,

Suppose I have a New Keynesian model in which the underlying shocks are TFP shocks.
Suppose I want to see the effect of "shock to labor supply" on GDP. Now, the shock is in quotation marks because the true underlying shocks are only to TFP, it's just I really want to graphically see through IRF, how decrease in labor affects GDP.

Can I add in dynare mod file, a phantom shock to the labour definition? The phantom shock is uncorrelated with true shocks to TFP.
Thus I can use IRF of the phantom shock to obtain the graphical IRF result I desire, which is, how decrease in labor affects GDP

Re: Phantom shock for IRF purpose

PostPosted: Mon Oct 20, 2014 5:14 am
by jpfeifer
Sorry, but I don't understand what you are trying to achieve. If you are using a first order approximation, you can add additional shocks to your model without affecting the IRFs to the other shocks already present. If you would subsequently do estimation, you would need to shut this shock off, which can be easily done.

If you are trying to add this shock to trace out a particular aspect of the response to a TFP shock, this most probably won't work because your two shocks act on very different margins.