Inferring Preferences in Multiple Criteria Decision Analysis Using a Logistic Regression Model

Publisher Website
Google Scholar
Abstract
A method is proposed for the analysis of multiple criteria decision making problems in an interactive environment, when decision-maker preferences are inconsistent with a simple utility model and/or are self-inconsistent (e.g., showing intransitivities). A maximum likelihood estimation procedure is invoked which is based on a logistic regression model relating the probability of selecting one decision option over another to a linear function of attribute values. The method is illustrated by application to multi-objective linear programming, where it serves as an alternative to the method of Zionts and Wallenius (Zionts, S., J. Wallenius. 1976. An interactive programming method for solving the multiple criteria problem. Management Sci. 22 652--663.), and allows for inconsistencies which are not satisfactorily handled in the Zionts-Wallenius approach.programming: multiple criteria, multi-attribute utility estimation, stochastic choice

This publication has 0 references indexed in Scilit: