Expected Subjective Value Theory (ESVT): A Representation of Decision Under Risk and Certainty

Abstract

We present a novel descriptive model of choice that achieves an efficient representation anchored to how the brain represents value. An individual's behavior is fully described by two primitives: an individual's "reward expectation'' and a free parameter we call "predisposition''. We demonstrate the breadth of the model, showing that it captures many of the behavioral phenomena described by Prospect Theory through a novel reformulation of its primitives. The model accounts for individual heterogeneity in behavior, employs fewer parameters than Prospect Theory, and retains neurobiological plausibility as a causal model of the choice process. Our theory makes a series of novel behavioral predictions amenable to future testing and includes novel explanations of the endowment effect and the observed heterogeneity in the probability weighting functions.

Keywords: utility, decision-making, reference point, neuroeconomics

JEL Classification: D03, D81, D87