Validation and Error in Multiplicative Utility Functions
Research rept. Oct 1977-Dec 1978
DECISIONS AND DESIGNS INC MCLEAN VA
Pagination or Media Count:
In this report an approach to the concept of error in utility assessment is proposed. Three components of error are considered and each component is related to four separate elicitation methods-all in the context of a general multiplicative multiattribute utility model. The methods are Keeney- Raiffa 1976 procedure, SMART Edwards, 1977, a social judgment theory SJT based regression model Hammond, Stewart, Brehmer and Steinmann, 1975 and a new method called Holistic Orthogonal Parameter Estimation or HOPE Barron and Person, 1978. If a general multiplicative model can be assumed to be an appropriate representation of the decision makers basic preference structure, error can occur in the direct estimation of the scaling constants and univariate utility functions for decomposition methods Keeney-Raiffa and SMART, or in the holistic assessments for holistic methods SJT and HOPE. Individual estimates may be merely noisy or may be fundamentally incorrect. Furthermore, the utility model may be incorrectly specified for example, an additive model, rather than a multiplicative model, may be used. The four assessment methods are considered in conjunction with errors of each kind.
- Statistics and Probability