I am attempting to augment a policy rule incorporating lending spreads (a la curdia woodford) where the policy rate falls with an increase in the lending spread, into the Gerali et al (2010) model however the Impulse Response Functions to a financial shock seem to be almost reversed: that is - incorporating a negative policy reaction to spreads seems to increase the policy rate when spreads widen.
I was wondering if anyone who was familiar with the Gerali model had any insights into the errors I was possibly making? I have a feeling it relates to the rational expectations of the banks in the model
Appreciate the time you have taken!
finshock.mod (28.1 KB)
I deleted Ale_e_Andre 's SPV part of the code and used it to estimate parameters, but the estimation is still not OK, even I modified the mode_compute=4, 5,6,7,8 9. the data file is your original vesion.
do you know what’s wrong?
GNSS_P3_estimation.mod (27 KB)
Looking at the estimation command: why did you put 1 for mh_replic? I put 100 as trial value, and I got results, the program run and we have IRFs. Anyway, you should probably increase the number of replications as in the original paper.
This is my first impression. Check your results in terms of both econometric and economic robustness.
Hope it is helpful.
Hello! I hope you had nice christmas holydays and are motivated to help me on this
I am also replicating the Gerali et al (2010) model. I have the code for the baseline model from the MMB2 model database. Now I am wondering if the authors change the code in some way before simulating the monetary policy shock. For the technology shock everything is fine but when I use the code to simulate a monetary policy shock, especially inflation and the policy rate (but also consumption, output and deposits) react too strongly. I attach my code. Thank you!
Banking_Code.mod (28.3 KB)