Title: Using the language of summarising statistics in dynamic decisions

Authors: Kurt J. Engemann, Holmes E. Miller, Ronald R. Yager

Addresses: Department of Information and Decision Technology Management, Iona College, New Rochelle, NY 10801, USA. ' Department of Business Administration, Muhlenberg College, Allentown, PA 18104, USA. ' Machine Intelligence Institute, Iona College, New Rochelle, NY 10801, USA

Abstract: Decision making under risk in a dynamic environment involves making a sequence of decisions which are interspersed with probabilistic events leading to an uncertain outcome. Decision makers typically use the expected value as a measure to summarise the value of a particular decision policy. We introduce a new methodology, using the language of summarising statistics, to allow the decision maker to assess the worth of a course of action. The decision maker may systematically align probabilistic elements with his own judgement. Summarising statistics provide descriptive measures to aid the decision maker|s understanding of a complex dynamic decision environment and to select a preferred sequence of decisions. We apply the Ordered Weighted Averaging operator as a basis for generating summarising statistics. We show how weight generating functions can induce different forms of Ordered Weighted Averaging operators which in turn revise event probabilities in the light of the decision maker|s disposition. We illustrate our methodology by determining a decision policy to counter cyberterrorism.

Keywords: bias; cyberterrorism; decision analysis; expected value; ordered weighted averaging operator; summarising statistics; weight generating functions.

DOI: 10.1504/IJTPM.2003.003986

International Journal of Technology, Policy and Management, 2003 Vol.3 No.3/4, pp.322 - 342

Published online: 10 May 2004 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article