Talk:Hyperbolic discounting

This article is within the scope of WikiProject Philosophy, a collaborative effort to improve the coverage of content related to philosophy on Wikipedia. If you would like to support the project, please visit the project page, where you can get more details on how you can help, and where you can join the general discussion about philosophy content on Wikipedia.PhilosophyWikipedia:WikiProject PhilosophyTemplate:WikiProject PhilosophyPhilosophy articles

Low

This article has been rated as Low-importance on the project's importance scale.

Associated task forces:
/
	Logic

Economics Low‑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
Low	This article has been rated as Low-importance on the project's importance scale.

Difficult wording[edit]

The sentence "Whether high rates of hyperbolic discounting precede addictions or vice-versa is currently unknown, although some studies have reported that high-rate discounting rats are more likely to consume alcohol[12] and cocaine[13]" was difficult to read. The first part of the sentence can probably be omitted. Correlation does not imply causation.

Agree with the comment below from two years ago. The graph is totally conflitant with the text. If Im wrong, please explain why. — Preceding unsigned comment added by 164.85.67.80 (talk) 16:30, 20 October 2011 (UTC)[reply]

The latter picture is very misleading compared to the text of this article. The point of hyperbolic discounting is that it overdiscounts in the short-term and underdiscounts in the long term as compared to exponential discounting. This is not expressed at all in the graph and in fact it appears more like the opposite

Sprigint (talk) 06:42, 20 December 2009 (UTC)[reply]

While it may be "dynamically inconsistent" to choose $50 now instead of $100 a year from now, but when given the choice between $50 in five years or $100 choose $100 in six years, it seems that it could be rational. The reason is simply that if the other party is around 5 years and then pays you, they would likely also be around 6 years and pay you, but that they are around 1 day does not make it very likely that they will be around 365 days.

Assuming rational choice and working backwards it would even be possible to back-calculate how people estimate the probabilities that another party will be around time t from now. I would guess a convex function similar to what you would get if you assumed that disappearance is a Poisson-process.

Filur 21:58, 6 November 2006 (UTC)[reply]

So-called "experimenter trust effects" have been examined in the literature on discount functions, and they don't suffice to explain the diverse range of observed evidence. However, you are very correct that in principle they could explain the example quoted above. Jeremy Tobacman 16:46, 17 February 2007 (UTC)[reply]

I know you guys know a lot about this stuff, but the intro paragraph needs to be reworked. As a person with casual interest in economics it doesn't even make sense. Maybe a simple editing would be fine.After reading the intro am I supposed to believe that hyperbolic discounting means as a human I prefer smaller payments sooner when I could get larger payments later? --69.140.59.32 16:37, 11 July 2007 (UTC)[reply]

agree with above. first paragraph is confusing. it may make sense to you but not to everyone else, which is kind of the point of the first paragraph. so it needs to be re-worked. —Preceding unsigned comment added by 84.13.249.253 (talk) 03:53, 25 December 2007 (UTC)[reply]

I deleted the claim that there is now evidence that quasi-hyperbolic preferences are more accurate. While there was a reference, the referenced paper does not actually prove this claim. —Preceding unsigned comment added by 98.207.93.141 (talk) 03:59, 15 May 2009 (UTC)[reply]

Pigeon studies[edit]

In the first section pigeon studies are mentioned: "In studies of pigeons, for example the pigeon is given two buttons: button A provides a small amount of food quickly while button B provides more seed but after a delay. The bird then experiments for a while and settles on preferring A or B."
But what does this say about hyperbolic discounting??? Please clarify, otherwise it should be removed. Lova Falk talk 17:02, 23 May 2010 (UTC)[reply]

I removed the section. The paragraph didn't even describe the results of the study, which would be essential to call it hyperbolic discounting. But even the setup described wouldn't produce results which would reveal preference profiles across time. The study just dealt with discounting in general it seems. In any case, even if the study was relevant, it was so tangential it shouldn't have been in the lead section in the first place. MarginalCost (talk) 18:40, 2 December 2010 (UTC)[reply]

Hyperbolic discounting paradigms couched in hypothetical situations ignore the probability of black swan events, why is that. For example, I am fifty years old and no one has ever walked up and said, hey would you like $100.00 today or $101.00 tomorrow? This black swan event has the present probability of 1/(365 days x 12 months x 50 years). Now to expect this black swan event to occur on two consecutive days would have a probability of 1/(365 x 12 x 50 x 2). The rational question is $100.00 today = a probability of 1, or waiting until tomorrow = 1/(365 x 12 x 50 x 2) = 1/438000 = p(0.000002). p = 1 vs. p = 0.000002?. Something seems to be missing from the conversation. — Preceding unsigned comment added by 69.47.135.131 (talk) 22:21, 6 March 2012 (UTC)[reply]

Discount rate[edit]

What does discount rate even mean in the context of hyperbolic discounting? According to the first source cited in the article, in the conventional discounted utility model, the discount factor is 1/(1+ρ), where ρ is the discount rate. Hyperbolic discounting means the discount rate in this expression decreases with time. The sentence "The rate depends on a variety of factors, including the species being observed, age, experience, and the amount of time needed to consume the reward." however implies that there is such a thing as a single discount rate in hyperbolic discounting. Can someone clarify? — Preceding unsigned comment added by Allion (talk • contribs) 21:40, 12 November 2012 (UTC) --Allion (talk) 23:02, 12 November 2012 (UTC)[reply]

Hyperbolic discounting does not have a constant discount rate but, as you correctly say, one that decreases with time. It is still true that the rate is affected by a variety of factors. MartinPoulter (talk) 16:31, 22 March 2014 (UTC)[reply]

Problem with current illustration[edit]

Unless parameters are used which show these curves crossing we don't see missing the main point here which is preference reversal. BO | Talk 01:20, 16 September 2013 (UTC)[reply]

I'm removing this graph, as it does not reflect the importance of hyperbolic discounting in the context of behavioral economics, and actually confuses the matter. If someone wants to create an appropriate illustration, it's important to fit the curve parameters to reflect the actual use of hyperbolic discounting by humans/animals. See for example, figures 1A and 1B at http://www.picoeconomics.org/HTarticles/Bkdn_Precis/Precis2.html Trevyn (talk) 08:56, 29 August 2017 (UTC)[reply]

Interactive graphs[edit]

I create interactive learning materials for economics as part of my job, and two are about this article's topic. I won't add them as external links myself because of conflict of interest, but see [1] and [2]. They are free works with a CC-BY licence. MartinPoulter (talk) 12:52, 15 March 2014 (UTC)[reply]

Prior distribution in Bayes example[edit]

How was the prior chosen? — Preceding unsigned comment added by Surement (talk • contribs) 04:00, 18 March 2016 (UTC)[reply]

Equation clarification[edit]

in the equation "(y/[(1 + r%)^n)" I don't understand the "/[" notation. Should it be "(y/[(1 + r%)^n])?" — Preceding unsigned comment added by Lakeadam (talk • contribs) 12:29, 31 March 2016 (UTC)[reply]

Step-by-step example[edit]

The "step-by-step example" subsection is not clear at all. I made some corrections to the formula, but it is still unclear how these calculations explain the time inconsistency. --Erel Segal (talk) 15:52, 15 November 2021 (UTC)[reply]