2018/07/28 by Christian Tarsney, Tarsney, Christian
Arts and Humanities · Decision Sciences · #Decision-Making and Behavioral Economics #Epistemology, Ethics, and Metaphysics #FOS: Economics and business #Philosophy and History of Science #Theoretical Economics (econ.TH)
paper · pdf · doi:10.48550/arxiv.1807.10895
openalex publication_date 2018/07/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The principle that rational agents should maximize expected utility or\nchoiceworthiness is intuitively plausible in many ordinary cases of\ndecision-making under uncertainty. But it is less plausible in cases of\nextreme, low-probability risk (like Pascal's Mugging), and intolerably\nparadoxical in cases like the St. Petersburg and Pasadena games. In this paper\nI show that, under certain conditions, stochastic dominance reasoning can\ncapture most of the plausible implications of expectational reasoning while\navoiding most of its pitfalls. Specifically, given sufficient background\nuncertainty about the choiceworthiness of one's options, many\nexpectation-maximizing gambles that do not stochastically dominate their\nalternatives "in a vacuum" become stochastically dominant in virtue of that\nbackground uncertainty. But, even under these conditions, stochastic dominance\nwill not require agents to accept options whose expectational superiority\ndepends on sufficiently small probabilities of extreme payoffs. The sort of\nbackground uncertainty on which these results depend looks unavoidable for any\nagent who measures the choiceworthiness of her options in part by the total\namount of value in the resulting world. At least for such agents, then,\nstochastic dominance offers a plausible general principle of choice under\nuncertainty that can explain more of the apparent rational constraints on such\nchoices than has previously been recognized.\n