Abstract
Previous attempts at determining the optimum strategy for playing the HI-LO sequential gaming routine relied heavily on decision analysis methodology. In contrast the approach described in this paper involves a stochastic dynamic programming (backward recursion) treatment. Details of the new procedure are presented, together with illustrative - and revealing - results for an existing test application.
| Original language | English |
|---|---|
| Title of host publication | 49th Annual Conference of the Operational Research Society 2007, OR49|49th Annu. Conf. Oper. Res. Soc. 2007, OR 2007 |
| Publisher | Operational Research Society |
| Pages | 66-78 |
| Number of pages | 12 |
| Publication status | Published - 2007 |
| Event | 49th Annual Conference of the Operational Research Society 2007, OR49 - Edinburgh Duration: 1 Jul 2007 → … |
Other
| Other | 49th Annual Conference of the Operational Research Society 2007, OR49 |
|---|---|
| City | Edinburgh |
| Period | 1/07/07 → … |
Keywords
- Decision analysis
- Dynamic programming
- Gaming
- Markov processes