Decision Analysis Working Paper Abstract
Archive
WP030024
|
Title: Attribute Dominance Utility
Authors: Ali Abbas
Stanford University and Ronald A. Howard
Stanford University
Date: October 2003
Status: Working Paper
This paper presents an analogy between joint cumulative probability distributions
and a class of multiattribute utility functions, which we call attribute dominance
utility. We show that any utility function can be decomposed into a linear
combination of attribute dominance utility functions that are easier to elicit.
This decomposition makes no assumptions about utility independence and allows
the application of widely used techniques of joint probability assessment,
such as marginal-conditional utility assessments and the method of copulas
to assess general multiattribute utility functions. We introduce the notion
of utility inference analogous to Bayes’ rule for probability inference and
provide a graphical representation of utility functions, which we call utility
diagrams.
Check back soon for a downloadable version of this paper.
Back to the Decision Analysis Working Paper Index
