Seed misbehavior with Dynare++
Created by: davidrpugh
My colleague and I obtain different results for 2nd and 3rd order impulse response function for a basic RBC model using the same seed. We both obtain identical results using the same seed on our individual computers, but results are inconsistent across computers. I run Dynare++ on Macbook running OSX 10.6; my colleague uses Windows 7.
Any thoughts? Would be happy to share the model file if interested...but probably better for someone to replicate the issue with a different model on another set of computers.