shade-econ / sequence-jacobian

A unified framework to solve and analyze heterogeneous-agent macro models.
MIT License
253 stars 149 forks source link

Using Sequence Space Algorithm in a deterministic rbc #19

Open mabde055 opened 11 months ago

mabde055 commented 11 months ago

Dear all,

I would like to use the abc model in the following link:

https://github.com/shade-econ/sequence-jacobian/tree/master/notebooks

But, I need to use in a deterministic case (i.e., without shocks in TFP). Additionally, I need to determine a target in one variable. For example, I would like the capital (K) to be higher in the next periods by 20% compared to its steady state. Afterwords, I need to see the effect of this target on the rest of macro variables like consumption. If there is a template for rbc model doing same task, it would be appreciated. Alternatively, how can I modify the codes in this provided link (I mean nbc.ipynb to achieve the target. Thank you very much

mrognlie commented 6 months ago

If you want to study transitional dynamics from an exogenously higher initial capital K, then one simple option would be to add a shock that scales the incoming K. Say, you would let Scale by the name of the shock, and then replace K(-1) in all the equations with Scale*K(-1). You'd then set the steady state value of Scale to be 1, and study the impulse response to a shock to Scale at date 0 using either impulse_linear or impulse_nonlinear.