The code is written in Python (version 3.10) and has been tested on Linux (Debian 12) and Windows 11. It is also expected to work on macOS, though not tested.
code/ Folder with the source code
output/
|- figures/ Folder where Python writes the figures
|- simulation_results/ Folder where Python saves intermediate results
|- tables/ Folder where Python writes the tables
The only data used in the paper are interest rates and CPI from
FRED. They are not stored in this archive, instead,
they are automatically downloaded when executing the file code/real_rate.ipynb.
An API key is required to
enable automatic data download from FRED. This key
should be saved to the environment variable FRED_API_KEY to successfully run the file
code/real_rate.ipynb.
| Program | Figure | File (folder output/figures) |
|---|---|---|
code/real_rate.ipynb |
1 | real_rate.pdf |
| Produced with Visio without code | 2 | eq_pi.png |
code/plot_stability.ipynb |
3 | stab1.pdf and stab2.pdf |
code/irf_param.ipynb |
4 | irf_param_Xpast_mean.pdf |
code/irf_state.ipynb |
5 | irf_state_Xpast_base.pdf |
code/irf_demand_param.ipynb |
6 | irf_demand_param.pdf |
| Program | Table | File (folder output/tables) |
|---|---|---|
code/precision.py |
E.1 | lee_nk.tex and lee_delta_lambda.tex |
| Program or folder | Description |
|---|---|
code/pricing.py |
The key optimality package (speculative channel) |
code/pricing_demand.py |
The key optimality package (global demand channel) |
code/tnr |
Computes a quadrature rule for a truncated normal distribution |
All files in the folder code/tnr come from John Burkardt's website,
and are licensed under the GNU Lesser General Public License (file code/tnr/LICENSE).
The calculations are executed in parallel using Numba. To control the number of threads,
modify the following lines in code/pricing.py and code/pricing_demand.py:
set_num_threads(X)By default, Numba uses all available threads.