A quantitative approach to choose among multiple mutually exclusive decisions: comparative expected utility theory

TL;DR

This paper introduces a new quantitative decision theory for selecting among multiple mutually exclusive options, incorporating individual preferences without relying on probability weighting, aiming for practical and rational decision-making.

Contribution

It develops a simple, normative, and quantitative framework that models individual understandings and feelings in mutually exclusive decision scenarios, extending classic expected utility theory.

Findings

The new theory accommodates different decision-makers' preferences.

It does not require probability weighting functions.

Results are rational and personalized for each decision-maker.

Abstract

Mutually exclusive decisions have been studied for decades. Many well-known decision theories have been defined to help people either to make rational decisions or to interpret people's behaviors, such as expected utility theory, regret theory, prospect theory, and so on. The paper argues that none of these decision theories are designed to provide practical, normative and quantitative approaches for multiple mutually exclusive decisions. Different decision-makers should naturally make different choices for the same decision question, as they have different understandings and feelings on the same possible outcomes.The author tries to capture the different understandings and feelings from different decision-makers, and model them into a quantitative decision evaluation process, which everyone could benefit from. The basic elements in classic expected utility theory are kept in the new decision theory, but the influences from mutually exclusive decisions will also be modeled into the evaluation process. This may sound like regret theory, but the new approach is designed to fit multiple mutually exclusive decision scenarios, and it does not require a definition of probability weighting function. The new theory is designed to be simple and straightforward to use, and the results are expected to be rational for each decision-maker.