README.rst

Continuous State Dynamic Programming via Nonexpansive Approximation

This page collects files and computer code for the paper Continuous State Dynamic Programming via Nonexpansive Approximation.

Publication Details

Continuous State Dynamic Programming via Nonexpansive Approximation
John Stachurski
Computational Economics, 31 (2), 141--160, 2008

Abstract

This paper studies fitted value iteration for continuous state numerical dynamic programming using nonexpansive function approximators. A number of approximation schemes are discussed. The main contribution is to provide error bounds for approximate optimal policies generated by the value iteration algorithm.