Skip to content

Fully JAX-compatible python implementation of the DC-EGM algorithm from Iskhakov, Jorgensen, Rust, and Schjerning (QE, 2017).

License

Notifications You must be signed in to change notification settings

OpenSourceEconomics/dcegm

Repository files navigation

DC-EGM

Continuous Integration Workflow Codecov pre-commit.ci status Black

Python implementation of the Discrete-Continuous Endogenous Grid Method (DC-EGM) for solving dynamic stochastic lifecycle models of continuous (e.g. consumption-savings) and additional discrete choices.

The solution algorithm employs an extension of the Fast Upper-Envelope Scan (FUES) from Dobrescu & Shanker (2022).

References

  1. Iskhakov, Jorgensen, Rust, & Schjerning (2017). The Endogenous Grid Method for Discrete-Continuous Dynamic Choice Models with (or without) Taste Shocks. Quantitative Economics
  2. Christopher D. Carroll (2006). The method of endogenous gridpoints for solving dynamic stochastic optimization problems. Economics Letters
  3. Loretti I. Dobrescu & Akshay Shanker (2022). Fast Upper-Envelope Scan for Discrete-Continuous Dynamic Programming.

About

Fully JAX-compatible python implementation of the DC-EGM algorithm from Iskhakov, Jorgensen, Rust, and Schjerning (QE, 2017).

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages