Article ID: | iaor1988723 |
Country: | United States |
Volume: | 37 |
Issue: | 2 |
Start Page Number: | 339 |
End Page Number: | 344 |
Publication Date: | Mar 1989 |
Journal: | Operations Research |
Authors: | Hopp Wallace J. |
Keywords: | markov processes |
A procedure for identifying forecast horizons in nonhomogeneous Markov decision processes, based on convergence results for relative value functions, is developed. Two different algorithmic implementations of this procedure are discussed, and a closed form expression for computing sufficiently long horizons to guarantee epsilon optimality is presented.