Hi all,
In a perfect foresight setup, I am trying to simulate a model where the CB implements QE while keeping the policy rate constant for X periods, i.e. pegging the interest rate to its steady state value for X periods, and then let it freely move according to a Taylor rule.
The problem is I don’t know how to implement such an experiment. Any intuition/suggestions would be highly appreciated.
Thanks
may be a starting point.
Thank you for your reply @jpfeifer .
I have used a shortcut for the implementation of the experiment that I am interested in.
In particular, if we assume a nonlinear Taylor rule with smoothing, and we raise the RHS to the power of “dummy”, where “dummy” is an exogenous variable, then we can shock that variable (dummy) with 0 for periods 1:4, and with 1 for periods 4:end.
If we do so, we indeed get a peg for the first 4 periods - so the nominal rate does not react to QE - and then reverts to a Taylor rule. However, the problem now becomes that Rank condition is violated (#unstable eigenval < #forward looking).
What are your thoughts about the shortcut that I took, and the new problem that arises?
That short cut may work. When does the error message appear? If you evaluate it at the dummy being 0, then the message is expected.