Open jstac opened 1 year ago
@jstac @JunnanZ Should we start moving towards GCY, or, how are we proceeding on this? Thanks!
Hi @Smit-create , yes, we can now start moving on this. Do you have time to take a shot, just paralleling the continuous state version of SSY?
The rules for updating the state are given at the top of https://github.com/jstac/sdfs_via_autodiff/blob/main/code/gcy/gcy_model.py
Perhaps it's worth it to code up the loglinear price-consumption ratio for the GCY model, to feed Newton's method as an initial guess. It might not work but it's worth a try.
How to compute it is detailed in Section A.2 of this paper. @Smit-create, do you think you can work this out? It should be similar to what we have done for the SSY model here: https://github.com/jstac/sdfs_via_autodiff/blob/main/code/ssy/ssy_model.py#L94.
Thanks for sharing the paper @JunnanZ. I can have a look and get back if I have any doubts.
@jstac , @JunnanZ do we have any other pending tasks for this issue or in general?
To be completed once we are happy with SSY continuous state version.