Paper #299

Title:

Solving higher-dimensional continuous time stochastic control problems by value function regression

Author:

Michael Reiter

Date:

March 1997 (Revised: June 1998)

Abstract:

The paper develops a method to solve higher-dimensional stochastic control problems in continuous time. A finite difference type approximation scheme is used on a coarse grid of low discrepancy points, while the value function at intermediate points is obtained by regression. The stability properties of the method are discussed, and applications are given to test problems of up to 10 dimensions. Accurate solutions to these problems can be obtained on a personal computer.

Keywords:

Dynamic Programming, stochastic control, approximation

JEL codes:

C61

Area of Research:

Microeconomics

Published in:

Journal of Economic Dynamics and Control, 23, (1999), pp. 1329-53

Download the paper in PDF format