Must do
-
Reorganize
- day 1 = Python (getting started, functions), NumPy (MCs, pass by ref), SciPy (equilib.)
- day 2 = Numba, JAX (nd equilib, european option, sS dynamics)
- day 3 = Fixed points, Newton for FPs, job search
- day 4 = DP theory, DP with JAX
-
Add asset pricing exercises to Markov chain notebook (or make separate notebook)
-
Integrate SS dynamics notebook
Might do
-
Prep Arellano JAX version?
-
Prep Elliot et al. default cascades?
-
Add cuda version of GPU programming to European option example, just to illustrate?
-
Discuss Leontief / IO with supply constraints (monotone concave operators)?
-
Add Toda example and solve by autodiff?
-
Test Anderson Acceleration with VFI?
-
R&D with IID costs as an example of dimensionality reduction?
-
Introduce NumPy via Leontief / IO?