Changing seed for calculating impulse responses
Posted: Sun Jul 10, 2011 4:00 pmHi,
I am working on Smets-Wouters (2007,AER) paper. Everytime I run the code, I get exactly the same magnitudes for deviations from the steady state. I understand that impulse response functions are solutions to the model rather than simulation. But the magnitudes of the steady state deviations are calculated for a random shock with the given standard deviation. Since the shock is random it must be generated from some random number generator. Getting the same deviations made me think that there must be some seed for those random shocks that are generated. But I can't find where the shocks for impulse responses are generated in the given .m files. I have tried set_dynare_seed but it works only for data simulation but not the impulse responses.
So is there a way to make sure the shocks generated for the IRFs are truly random in Dynare?
The .mod file is attached.
Thank you,
I am working on Smets-Wouters (2007,AER) paper. Everytime I run the code, I get exactly the same magnitudes for deviations from the steady state. I understand that impulse response functions are solutions to the model rather than simulation. But the magnitudes of the steady state deviations are calculated for a random shock with the given standard deviation. Since the shock is random it must be generated from some random number generator. Getting the same deviations made me think that there must be some seed for those random shocks that are generated. But I can't find where the shocks for impulse responses are generated in the given .m files. I have tried set_dynare_seed but it works only for data simulation but not the impulse responses.
So is there a way to make sure the shocks generated for the IRFs are truly random in Dynare?
The .mod file is attached.
Thank you,