Dear all,
I would like to estimate a New Keynesian DSGE model with zero lower bound constraint. Could you let me know if this is currently feasible in Dynare?
Many thanks!
No, it’s not. You need some specialized estimation algorithm for that.
Dear mcrel,
If you are interested, https://www.aeaweb.org/articles?id=10.1257/mac.20170269
Anzoategui et al. (2019) use data before the ZLB starts binding in the United States to estimate the structural parameters of their model, but do use data during the period characterised by a binding ZLB to indentify shocks and other latent objects in their model. To do this, they rely on Occbin code.
Should this approach of theirs suit your purpose, their replication material (available at the same link) contains the code they have used to implement their algorithm.
Furthermore, if you are thinking about estimating a model with a binding ZLB in your estimation sample, I suggest you have a look at another very novel paper by Atkinson et al. (2019) https://www.sciencedirect.com/science/article/pii/S0304393219301175