Talk:Expected utility hypothesis

Game theory Top‑importance

	This article is part of WikiProject Game theory, an attempt to improve, grow, and standardize Wikipedia's articles related to Game theory. We need your help! Join in \| Fix a red link \| Add content \| Weigh inGame theoryWikipedia:WikiProject Game theoryTemplate:WikiProject Game theorygame theory articles
Top	This article has been rated as Top-importance on the importance scale.

Economics High‑importance

	Business and economics portal This article is within the scope of WikiProject Economics, a collaborative effort to improve the coverage of Economics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.EconomicsWikipedia:WikiProject EconomicsTemplate:WikiProject EconomicsEconomics articles
High	This article has been rated as High-importance on the project's importance scale.

Statistics Low‑importance

	This article is within the scope of WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.StatisticsWikipedia:WikiProject StatisticsTemplate:WikiProject StatisticsStatistics articles
Low	This article has been rated as Low-importance on the importance scale.

Effective Altruism Mid‑importance

	This article is within the scope of WikiProject Effective Altruism, a collaborative effort to improve the coverage of topics relevant to effective altruism on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Effective AltruismWikipedia:WikiProject Effective AltruismTemplate:WikiProject Effective AltruismEffective Altruism articles
Mid	This article has been rated as Mid-importance on the importance scale.

Archives

Archive 1

Archive 2

This page has archives. Sections older than 365 days may be automatically archived by when more than 5 sections are present.

__DTSUBSCRIBEBUTTONDESKTOP__{"headingLevel":2,"name":"h-Rieger-2018-04-03T14:02:00.000Z","type":"heading","level":0,"id":"h-Suggestion_for_improvement-2018-04-03T14:02:00.000Z","replies":["c-Rieger-2018-04-03T14:02:00.000Z-Suggestion_for_improvement","c-Editeur24-2020-11-23T21:01:00.000Z-Suggestion_for_improvement"],"text":"Suggestion for improvement","linkableTitle":"Suggestion for improvement"}-->

Suggestion for improvement

__DTELLIPSISBUTTON__{"threadItem":{"headingLevel":2,"name":"h-Rieger-2018-04-03T14:02:00.000Z","type":"heading","level":0,"id":"h-Suggestion_for_improvement-2018-04-03T14:02:00.000Z","replies":["c-Rieger-2018-04-03T14:02:00.000Z-Suggestion_for_improvement","c-Editeur24-2020-11-23T21:01:00.000Z-Suggestion_for_improvement"]}}-->

__DTLATESTCOMMENTTHREAD__{"id":"c-Editeur24-2020-11-23T21:01:00.000Z-Suggestion_for_improvement","timestamp":"2020-11-23T21:01:00.000Z"}__-->__DTCOMMENTCOUNT__2__-->__DTAUTHORCOUNT__2__-->

__DTSUBSCRIBEBUTTONMOBILE__{"headingLevel":2,"name":"h-Rieger-2018-04-03T14:02:00.000Z","type":"heading","level":0,"id":"h-Suggestion_for_improvement-2018-04-03T14:02:00.000Z","replies":["c-Rieger-2018-04-03T14:02:00.000Z-Suggestion_for_improvement","c-Editeur24-2020-11-23T21:01:00.000Z-Suggestion_for_improvement"],"text":"Suggestion for improvement","linkableTitle":"Suggestion for improvement"}-->

What is really missing on this page is a real summary of what this theory is at the beginning (instead it starts with explaining what is risk aversion) and the central formula E[u(X)]=p1 u(x1)+p2 u(x2)+... Maybe somebody has time to work this into the article?

(I myself published some research in decision theory, so I probably could improve the article by myself, but I am quite busy now and before I postpone forever, I'd rather ask for help...)

Rieger (talk) 14:02, 3 April 2018 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"2018-04-03T14:02:00.000Z","author":"Rieger","type":"comment","level":1,"id":"c-Rieger-2018-04-03T14:02:00.000Z-Suggestion_for_improvement","replies":[]}}-->

I just did this, with a numerical example in the introduction. The page still heads into history and critique before really explaining expected utility-- the central formula is down quite a ways by now, and even risk aversion doesn't appear for a while.

I suggest rearranging the page drastically, with explanations and development (risk aversion, etc.) right after the introduction, then history, then criticisms. The criticism sections, which occupy most of the page, should perhaps be moved to a separate page with just a summary here. Or, they could be linked together more--- the page has probably been edited so much by different people that it doesn't hang together very well.

editeur24 (talk) 21:01, 23 November 2020 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"2020-11-23T21:01:00.000Z","author":"Editeur24","type":"comment","level":1,"id":"c-Editeur24-2020-11-23T21:01:00.000Z-Suggestion_for_improvement","replies":[]}}-->

__DTSUBSCRIBEBUTTONDESKTOP__{"headingLevel":2,"name":"h-Editeur24-2020-11-23T21:07:00.000Z","type":"heading","level":0,"id":"h-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped-2020-11-23T21:07:00.000Z","replies":["c-Editeur24-2020-11-23T21:07:00.000Z-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped"],"text":"The section \"Savage's Representation Theorem\" should be fixed or dropped","linkableTitle":"The section \"Savage's Representation Theorem\" should be fixed or dropped"}-->

The section "Savage's Representation Theorem" should be fixed or dropped

__DTELLIPSISBUTTON__{"threadItem":{"headingLevel":2,"name":"h-Editeur24-2020-11-23T21:07:00.000Z","type":"heading","level":0,"id":"h-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped-2020-11-23T21:07:00.000Z","replies":["c-Editeur24-2020-11-23T21:07:00.000Z-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped"]}}-->

__DTLATESTCOMMENTTHREAD__{"id":"c-Editeur24-2020-11-23T21:07:00.000Z-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped","timestamp":"2020-11-23T21:07:00.000Z"}__-->__DTCOMMENTCOUNT__1__-->__DTAUTHORCOUNT__1__-->

__DTSUBSCRIBEBUTTONMOBILE__{"headingLevel":2,"name":"h-Editeur24-2020-11-23T21:07:00.000Z","type":"heading","level":0,"id":"h-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped-2020-11-23T21:07:00.000Z","replies":["c-Editeur24-2020-11-23T21:07:00.000Z-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped"],"text":"The section \"Savage's Representation Theorem\" should be fixed or dropped","linkableTitle":"The section \"Savage's Representation Theorem\" should be fixed or dropped"}-->

The section "Savage's Representation Theorem" should be fixed or dropped. It does not say what assumptions P1-P7 are nor explain the notation, so it is too incomplete to be useful. Shall I delete the section? --editeur24 (talk) 21:07, 23 November 2020 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"2020-11-23T21:07:00.000Z","author":"Editeur24","type":"comment","level":1,"id":"c-Editeur24-2020-11-23T21:07:00.000Z-The_section_\"Savage's_Representation_Theorem\"_should_be_fixed_or_dropped","replies":[]}}-->

__DTSUBSCRIBEBUTTONDESKTOP__{"headingLevel":2,"name":"h-Michael_Hardy-2021-09-15T05:48:00.000Z","type":"heading","level":0,"id":"h-Puzzling_edit-2021-09-15T05:48:00.000Z","replies":["c-Michael_Hardy-2021-09-15T05:48:00.000Z-Puzzling_edit"],"text":"Puzzling edit","linkableTitle":"Puzzling edit"}-->

Puzzling edit

__DTELLIPSISBUTTON__{"threadItem":{"headingLevel":2,"name":"h-Michael_Hardy-2021-09-15T05:48:00.000Z","type":"heading","level":0,"id":"h-Puzzling_edit-2021-09-15T05:48:00.000Z","replies":["c-Michael_Hardy-2021-09-15T05:48:00.000Z-Puzzling_edit"]}}-->

__DTLATESTCOMMENTTHREAD__{"id":"c-Michael_Hardy-2021-09-15T05:48:00.000Z-Puzzling_edit","timestamp":"2021-09-15T05:48:00.000Z"}__-->__DTCOMMENTCOUNT__1__-->__DTAUTHORCOUNT__1__-->

__DTSUBSCRIBEBUTTONMOBILE__{"headingLevel":2,"name":"h-Michael_Hardy-2021-09-15T05:48:00.000Z","type":"heading","level":0,"id":"h-Puzzling_edit-2021-09-15T05:48:00.000Z","replies":["c-Michael_Hardy-2021-09-15T05:48:00.000Z-Puzzling_edit"],"text":"Puzzling edit","linkableTitle":"Puzzling edit"}-->

This edit is puzzling. It uses some very odd notation, thus:

f < g ⇐⇒ Z Ω u(f(ω)) dP ≥ Z Ω u(g(ω)) dP

My best guess is that this was intended:

f<g\Longleftrightarrow \int _{\Omega }u(f(\omega ))\,dP\geq \int _{\Omega }u(g(\omega )\,dP.

However, the cited work by Li, Loomes, and Pogrebna has nothing about this. Michael Hardy (talk) 05:48, 15 September 2021 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"2021-09-15T05:48:00.000Z","author":"Michael Hardy","type":"comment","level":1,"id":"c-Michael_Hardy-2021-09-15T05:48:00.000Z-Puzzling_edit","replies":[]}}-->

__DTSUBSCRIBEBUTTONDESKTOP__{"headingLevel":2,"name":"h-MaxEnt-20230927141100","type":"heading","level":0,"id":"h-Rewrite_lead_sentences-20230927141100","replies":["c-MaxEnt-20230927141100-Rewrite_lead_sentences"],"text":"Rewrite lead sentences","linkableTitle":"Rewrite lead sentences"}-->

Rewrite lead sentences

__DTELLIPSISBUTTON__{"threadItem":{"headingLevel":2,"name":"h-MaxEnt-20230927141100","type":"heading","level":0,"id":"h-Rewrite_lead_sentences-20230927141100","replies":["c-MaxEnt-20230927141100-Rewrite_lead_sentences"]}}-->

__DTLATESTCOMMENTTHREAD__{"id":"c-MartinPoulter-20231126113600-MaxEnt-20230927141100","timestamp":"20231126113600"}__-->__DTCOMMENTCOUNT__2__-->__DTAUTHORCOUNT__2__-->

__DTSUBSCRIBEBUTTONMOBILE__{"headingLevel":2,"name":"h-MaxEnt-20230927141100","type":"heading","level":0,"id":"h-Rewrite_lead_sentences-20230927141100","replies":["c-MaxEnt-20230927141100-Rewrite_lead_sentences"],"text":"Rewrite lead sentences","linkableTitle":"Rewrite lead sentences"}-->

I'm not a trained economist, but you'd be hard pressed to find an armchair more often steeped in the public discourse. The existing lead contained allusive, undefined language, so I revised it as follows:

The expected utility hypothesis is a foundational assumption in mathematical economics concerning human preference when decision making under uncertainty. It postulates that a rational agent maximizes utility, as formulated in the mathematics of game theory, based on their risk aversion. Rational choice theory, a cornerstone of microeconomics, builds upon the expected utility of individuals to model aggregate social behaviour.

To my eye, to a 90% standard, this is clear, cogent, and suitably framed within the discipline. But I can't be certain it's entirely in pedagogical alignment with how professionals view the matter. Please review and revise accordingly. — MaxEnt 14:11, 27 September 2023 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"20230927141100","author":"MaxEnt","type":"comment","level":1,"id":"c-MaxEnt-20230927141100-Rewrite_lead_sentences","replies":["c-MartinPoulter-20231126113600-MaxEnt-20230927141100"]}}-->

I think this was an improvement over what was there, and I've improved it further to fix these problems:

"based on their risk aversion" Risk aversion can't be modelled in expected utility theory, so to suggest it is somehow essential to the theory is going to mislead readers.
"as formulated in the mathematics of game theory" Expected utility theory applies to more situations than game theory does, so if readers think that game theory is essential to expected utility theory, they will get the wrong impression That expected utility came out of analysis of games is a historical detail which isn't needed to understand the theory.
"maximizes utility" with a link to the Utilitarianism article: this is the most problematic bit. English speakers who don't already know the topic will be familiar with "expectation" and "hypothesis", but the special meaning of "utility" - as a numerical measure of subjective desirability of a state or outcome - is going to be most crucial to them understanding the topic of the article. So when they encounter the unfamiliar term, it's understandable if they click the link. The Utilitarianism article is not the place to go to learn what utility is: it just discourages them by presenting them with a lot more unfamiliar concepts. So, if clarity is the goal, it's better without that wiki-link.
"concerning human preference when decision making under uncertainty" Apart from the dodgy grammar, there are semantic problems. "Human" isn't strictly correct as a definition because other kinds of decision making (organisational) are in the scope of expected utility. Preferences are usually treated as a more foundational concept than utility is (for example in representation theorems), so it's not strictly true to say that utility maximisiation is a foundational assumption concerning preference.

This explains the changes I've just made to the lead. Definitely further improvement is possible! MartinPoulter (talk) 11:36, 26 November 2023 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"20231126113600","author":"MartinPoulter","type":"comment","level":2,"id":"c-MartinPoulter-20231126113600-MaxEnt-20230927141100","replies":[]}}-->

__DTSUBSCRIBEBUTTONDESKTOP__{"headingLevel":2,"name":"h-MartinPoulter-20231126150700","type":"heading","level":0,"id":"h-\"Criticism\"-20231126150700","replies":["c-MartinPoulter-20231126150700-\"Criticism\""],"text":"\"Criticism\"","linkableTitle":"\"Criticism\""}-->

"Criticism"

__DTELLIPSISBUTTON__{"threadItem":{"headingLevel":2,"name":"h-MartinPoulter-20231126150700","type":"heading","level":0,"id":"h-\"Criticism\"-20231126150700","replies":["c-MartinPoulter-20231126150700-\"Criticism\""]}}-->

__DTLATESTCOMMENTTHREAD__{"id":"c-Closed_Limelike_Curves-20240720212000-MartinPoulter-20231126150700","timestamp":"20240720212000"}__-->__DTCOMMENTCOUNT__2__-->__DTAUTHORCOUNT__2__-->

__DTSUBSCRIBEBUTTONMOBILE__{"headingLevel":2,"name":"h-MartinPoulter-20231126150700","type":"heading","level":0,"id":"h-\"Criticism\"-20231126150700","replies":["c-MartinPoulter-20231126150700-\"Criticism\""],"text":"\"Criticism\"","linkableTitle":"\"Criticism\""}-->

I think this section is mis-titled. The Tversky and Kahneman research that is summarised accepts the assumption that rational agents maximise subjective expected utility. Instead they say that people's behaviour doesn't follow SEU theory and so people choose irrationally (or that SEU isn't an appropriate normative standard). So it's confusing to label this a "criticism" of expected utility. Technically it's a criticism of the descriptive application of the expected utility hypothesis, rather than its normative application. There are different consistent positions one could take, including:

SEU describes what rational agents do, and - most of the time / well enough - what humans do: the "Homo Oeconomicus" idea.
SEU describes what rational agents do but not what human beings do: Tversky and Kahneman and a lot of subsequent psychological research.
SEU doesn't describe rational agents or human beings.
If you have total freedom in how you set utility functions and probability functions, then SEU can be fit to any behaviour, thus it doesn't say anything about rational versus irrational (I think this has come up as an option in John Broome's work).

For now I'm just raising this issue for discussion. I don't know what the solution is; it has to not just involve a change of heading but further explanation. MartinPoulter (talk) 15:07, 26 November 2023 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"20231126150700","author":"MartinPoulter","type":"comment","level":1,"id":"c-MartinPoulter-20231126150700-\"Criticism\"","replies":["c-Closed_Limelike_Curves-20240720212000-MartinPoulter-20231126150700"]}}-->

@MartinPoulter the issue with these articles (Expected utility hypothesis, rational choice theory, and decision theory) is that none of them distinguish between the terms as a prescriptive vs. descriptive model of rationality.

These articles need to be regrouped and restructured into two articles:

1. One titled "decision theory" or "rational choice" explaining them as prescriptive models of rational decision-making, and

2. One titled something like "rational choice theory" describing how these can be used as models of actual human behavior. Closed Limelike Curves (talk) 21:20, 20 July 2024 (UTC)[reply]__DTELLIPSISBUTTON__{"threadItem":{"timestamp":"20240720212000","author":"Closed Limelike Curves","type":"comment","level":2,"id":"c-Closed_Limelike_Curves-20240720212000-MartinPoulter-20231126150700","replies":[]}}-->