Size of a shock in Gali_2015_chapter_3_nonlinear.mod

General DSGE Modeling Replication Attempts
1

Hi,
the size of a monetary shock in the Model Gali_2015_chapter_3_nonlinear.mod of a stderr. of 0.0025^2 is interpreted in Basispoint per Quartal. Am i right that the stderr. is squared to ensure that it is a positive shock? What does this mean exacly?
And how do you interpret a similar size of a technology shock, since the technology shock is given as Log(A)?

Thanks a lot!

2

Read the text above the mod-file. The standard deviation is 0.025 so if including the variance, this number is squared, that is all.

3

Also note that this is a quarterly interest rate. So 0.0025 is 1 percent annualized. Regarding TFP: having something in logs means it is in percent as well.

1 Like
4

Thanks!
Basically that means talking about the same value of 0.0025^2 for a technology shock you could speak of a 0,25 percentage points increase for the technology level per Quater?!
Another question regarding the irfs that are produced:
Since inflation, nominal and real interest rate are annualized does that mean one Period on the x-achses is seen to be a year for those variables while for the ones that are not annualized one Period on the x-achses represents only 1/4 of a year?

5
  1. Yes, that is correct for TFP.
  2. No, the annualization only alters the interpretation of the variables’ values on the y-axis, not the x-axis. It’s still the response of the annualized rates after 1,2,3,… quarters.
1 Like
6

Alright thanks! :slightly_smiling_face: