Dear all,
I am currently recalibrating the DSGE model while using data different from the one in the original paper.
I was able to narrow down the steady-state system of equations to just few. However, is makes the whole setup really sensitive to the initial guess. I was wondering, is there a “smart” way to update the initial guess for the steady state solver while calibrating? I have tried persistent X0 approach, but I suppose it drives the initial guess too far away while constantly updating and cannot return back if needed.
Any suggestion would be highly appreciated.