David R. Pugh bio photo

David R. Pugh

A Data Scientist at KAUST Core Labs and Associate Fellow with the Institute of New Economics Thinking (INET) at University of Oxford.

Email Twitter LinkedIn Github Stackoverflow

Numerical Method for Economists

Numerical Methods for Economists is a suite of Python-based lab sessions covering many of the basic numerical techniques used to solve and simulate non-linear economic models. These hands-on labs are taught as part of the core graduate macroeconomics sequence for the Scottish Graduate Programme in Economics (SGPE). Partial funded for development and teaching of the course has been provided by the Scottish Insitute for Research in Economics (SIRE).

Topics currently covered

  • Finite-difference methods for solving ordinary differential equations (methods for solving both initial and boundary value problems); students are taught to apply these techniques by solving, simulating, and calibrating seminal models of economic growth.
  • Numerical methods for solving systems of non-linear equations; students are taught to apply these methods to solve for the equilibrium of search and match models of unemployment.
  • Perturbation methods; students will learn to apply these methods by using them to solve and simulate RBC and NK DSGE models.
  • Numerical methods for dynamic programming (including discretization, value function iteration, and policy function iteration). Students will learn to apply these techniques by applying them to deterministic and stochastic optimal growth models.

The emphasis of the labs is on the practical application of the methods and not on the mathematical theory underlying the techniques.


Below are some links to some of the more well developed notebooks used in past labs.

The entire set of course materials is hosted on GitHub.