Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Tuesday May 19 2020, @05:46AM   Printer-friendly
from the do-you-feel-lucky? dept.

[20200519_114228 UTC: Updated to remove possible loss as my perceived implication of the original question.--martyb]

If you were given $1,000 to play a game, would you accept a 50 percent chance to double your money or a 100 percent guarantee of gaining an additional $500?

Implied in the question was that a 50% chance to double the $1,000 was also a 50% chance to lose all of the $1,000. Put that way, I'd take the 100% guarantee of gaining $500 more. Hmm. But why did I make that choice? What if I started with just $10? Or even $1? Would I choose differently? What if I started with $100,000 or even $1,000,000? Then what would my choice be — and why?

That opening question was one of 17 hypotheticals posed when attempting to replicate 1979 foundational research on loss aversion and prospect theory.

Global Study Confirms Influential Theory Behind Loss Aversion:

A new global study offers a powerful confirmation of one of the most influential frameworks in all of behavioral sciences and behavioral economics: prospect theory, which when introduced in 1979 led to a sea change in understanding the irrational and paradoxical ways individuals make decisions and interpret risk with major impacts for science, policy, and industry. Led by a Columbia University Mailman School of Public Health researcher, the new study in 19 countries and 13 languages replicates the original study that provided the empirical basis for prospect theory. Results appear in Nature Human Behaviour.

Developed by Nobel Prize winner Daniel Kahneman and Amos Tversky, prospect theory has been called the most influential theoretical framework in all of the social sciences and popularized the concept of loss aversion, which says that people prefer small guaranteed outcomes over larger risky outcomes. The 1979 paper that launched the theory is today the most cited paper in economics and is among the most cited in psychological science.

The new study led by Kai Ruggeri, PhD, assistant professor of health policy and management, is a robust test of prospect theory at a scale commensurate with its impact—and the first to test the theory in so many countries, languages, currencies, and to focus on the generalizability of the theory. Ruggeri and colleagues used nearly identical methods to those in the original study, modifying them only to make currency values relevant for a 2019 sample within each country. [...] In all, 4,098 respondents who completed all the questions were included in the final analysis.

Results of 1979 study—now confirmed in the new global study—gave rise to prospect theory and upended orthodoxies around rational choices. Among the original study's findings: people tend to be risk-seeking when maximizing gains, but risk-averse when minimizing losses; our preferences may change depending on how they are rendered sequentially; and we tend to overweight small probabilities.

The researchers found that Kahneman and Tversky's 1979 empirical foundation for proposing prospect theory broadly replicates in all the countries they studied: they report a 90 percent replication in areas directly testing the theoretical contrasts at the heart of prospect theory.

Journal Reference:
Kai Ruggeri, Sonia Alí, Mari Louise Berge, et al. Replicating patterns of prospect theory for decision under risk, Nature Human Behaviour (DOI: 10.1038/s41562-020-0886-x)


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: 2) by The Mighty Buzzard on Tuesday May 19 2020, @10:33AM (15 children)

    Nope, the second one is only better on average and you can't average one data point. So, from an individual standpoint the first one is better but comes with inherent risk.

    --
    My rights don't end where your fear begins.
    Starting Score:    1  point
    Karma-Bonus Modifier   +1  

    Total Score:   2  
  • (Score: 4, Interesting) by Immerman on Tuesday May 19 2020, @01:01PM (8 children)

    by Immerman (3985) on Tuesday May 19 2020, @01:01PM (#996306)

    The entire point of expected value is to apply statistical properties to individual choices. No, you can't average a single data point, but before making the choice you have *zero* data points, and only a probability distribution to navigate.

    If you're only considering the possible good outcome, then you'd still choose the gamble if it were a 1% chance of $1000, and 99% chance of nothing. But very few people would take that choice, because they realize they'll almost certainly lose - the expected value of the gamble is only $10, while the expected value of the $500 is $500.

    It's true that you only get one shot at that particular choice, so the average outcome isn't directly relevant, but life is full of choices, and you don't know what the actual outcome of any of them will be - only a (usually very) rough probability distribution. If you regularly make choices based only on the best possible outcome, then you'll probably crash and burn early and often, because the best possible outcome is usually very unlikely. Over the course of a lifetime you will get something approximating the sum of the expected values of all your choices.

    • (Score: 2) by theluggage on Tuesday May 19 2020, @01:41PM (1 child)

      by theluggage (1797) on Tuesday May 19 2020, @01:41PM (#996338)

      If you're only considering the possible good outcome, then you'd still choose the gamble if it were a 1% chance of $1000, and 99% chance of nothing.

      That's the point - the relevance of an 'expected value' depends on the magnitude of the probability involved.

      For a ~ 50% chance, you'd have to play a lot of games before the average return converged on $500.

      For a ~0% or ~100% chance then you can be reasonably confident of hitting the average on the first go...

      It's true that you only get one shot at that particular choice, so the average outcome isn't directly relevant, but life is full of choices,

      ...but not choices with the same probabilities, rules and consequences. The way you think (and should think) about a 1% gamble is completely different to the 50/50 scenario.

      Anyway probability is not risk - the risk depends both on the probability and the magnitude of the consequences: You might reasonably take a 1% chance of losing $10 in a game, but 1% chance of death is probably a bit on the high side for most people...

      There's a lot of that going around at the moment - and one problem is the conflict between the (for most people) very small personal risk of catching COVID and dying vs. the risk to a hospital service serving a population of millions getting swamped by even a modest percentage increase in sickness.

      • (Score: 2) by Immerman on Wednesday May 20 2020, @03:13PM

        by Immerman (3985) on Wednesday May 20 2020, @03:13PM (#996919)

        >That's the point - the relevance of an 'expected value' depends on the magnitude of the probability involved.

        That depends mostly on how easily you can absorb the risk. Because you're right - risk is a very different concept

        If you're gambling with pocket change - very low risk - then it's pretty much worth it to take any bet where the buy-in is even slightly lower than the expected value - over the course of many such bets (no two the same) you'll almost certainly come out ahead. Of course the expected value is determined by the probability: If you're gambling for $1000:
        - and the probability is 50%, then the expected value is $500
        - and the probability is 1%, then the expected value is $10,
        - and the probability is 100%, then the expected value is $990

        On the other hand, if the associated risk of losing is high - e.g. buying in means that if you lose you can't pay your mortgage and lose your house... then you should really be factoring that risk into the expected cost: it's not 50% chance of winning $500 ahead versus 50% chance of losing $500 - it's versus a 50% chance of losing $500 *and your house*. Call the house $100k, and instead of the expected value of that bet being $0, the expected value *for you* would be -$50k. You'd need an almost guaranteed win to expect to break even.

        As for putting your life on the line... expected value is really a financial concept, though I suppose if you can comfortably assign a dollar value to your life it's still possible. Governments do so routinely.

    • (Score: 2) by https on Tuesday May 19 2020, @11:10PM

      by https (5248) on Tuesday May 19 2020, @11:10PM (#996615) Journal

      If you regularly make choices based only on the best possible outcome, then you'll probably crash and burn early and often, because the best possible outcome is usually very unlikely. Over the course of a lifetime you will get something approximating the sum of the expected values of all your choices.

      ...except all the people betting safely have actually acquired something in the interim, leaving you begging for scraps and fuel most of the time, and probably a lot of debt and burned bridges along the way. Loser.

      --
      Offended and laughing about it.
    • (Score: 2) by The Mighty Buzzard on Wednesday May 20 2020, @02:30AM (4 children)

      Over the course of a lifetime you will get something approximating the sum of the expected values of all your choices.

      You'd think that at first glance but you'd be horribly wrong. Ask any investment advisor and they'll tell you that you need to be taking risky investments in your 20s and 30s and more and more conservative positions the closer you get to retirement. That's because a win early on can be compounded quite a lot over time while a loss early on is only going to hurt equivalent of its face value.

      --
      My rights don't end where your fear begins.
      • (Score: 2) by rondon on Wednesday May 20 2020, @03:15AM (1 child)

        by rondon (5167) on Wednesday May 20 2020, @03:15AM (#996703)

        I... your last statement is very, very wrong. Why did you just completely ignore the time value of money? What fucking world do we live in where gaining money has extra value but losing money does not?

        The investment advisor is telling you to make risky investments because you have more time for the fuck-ups to average out to a higher total return, not because a big gain compounds better than a small one.

        • (Score: 2) by The Mighty Buzzard on Wednesday May 20 2020, @03:33AM

          A big gain early on leaves you with more to compound over time. Early money earns more than late money by a damned lot. A loss early on loses you a pittance because you don't have much to lose. Early losses can be recovered from, late losses have no such guarantee.

          --
          My rights don't end where your fear begins.
      • (Score: 2) by Immerman on Wednesday May 20 2020, @03:41PM (1 child)

        by Immerman (3985) on Wednesday May 20 2020, @03:41PM (#996934)

        What you're ignoring, is that you haven't only lost the money today, but *also* all the money that that lost money would have compounded into.

        If you're counting a $100 win today as winning the $300 it will eventually grow in to, then you *also* need to count a $100 loss as losing the $300 that it would have eventually grown into.

        It's still worth taking higher risks when you're young, just not for those reasons. If you're doing it right the higher risk gambles will have a higher expected (probability-adjusted) return on investment, and it makes good financial sense to take that gamble for as long as losing doesn't carry any serious externalized risks. Making lower risk investments as you age is a concession to the fact that the externalized risks of losing increase as you get closer to needing to cash out.

        If you're *not* doing it right - e.g. you're making riskier investments whose probability-adjusted expected return on investment isn't higher than the lower-risk investments... then I would be suspicious of the advisor suggesting such a thing. Either they're incompetent, drinking the kool-aid, or intentionally misleading you for their own profit. They are after all in a position to win when you win, and walk away when you lose.

        • (Score: 2) by The Mighty Buzzard on Wednesday May 20 2020, @11:43PM

          See, there's a couple fundamental misconceptions you seem to have about gambling of any sort. And rest assured playing the markets is gambling. You're just betting on the competence of people rather than horses. First, you never count on winning. Second, you never use money that's going to hurt if you lose it all.

          --
          My rights don't end where your fear begins.
  • (Score: 0) by Anonymous Coward on Tuesday May 19 2020, @01:05PM (1 child)

    by Anonymous Coward on Tuesday May 19 2020, @01:05PM (#996308)

    On average the two are equal, just the one maximizes risk, while the other minimizes it.

    That is your choice to take the expected value with no risk, or take double that value with equal chance of getting nothing.

  • (Score: 0) by Anonymous Coward on Tuesday May 19 2020, @02:53PM (3 children)

    by Anonymous Coward on Tuesday May 19 2020, @02:53PM (#996364)

    So, you play the lottery, don't you? Same principle you are exposing, isn't it?

    • (Score: 2) by The Mighty Buzzard on Wednesday May 20 2020, @02:32AM (2 children)

      Newp. I don't gamble. And I wasn't advocating it, simply stating a fact.

      --
      My rights don't end where your fear begins.
      • (Score: 0) by Anonymous Coward on Wednesday May 20 2020, @03:44AM (1 child)

        by Anonymous Coward on Wednesday May 20 2020, @03:44AM (#996730)

        You're problem here is that you started by assigning a value judgment to one outcome. "One is better but with risk" is not a a proper valuation of the two scenarios. While I agree with your general point, you are ignoring everyone else and just re-stating your position with no regard to the specifics. Just one of the reasons you get a lot of flak around here, you should be more open to discussing ideas instead of setting up a foxhole.

        • (Score: 2) by The Mighty Buzzard on Wednesday May 20 2020, @03:52AM

          Blah blah blah. If you want to look at it a different way, go for it. But if you think I'm going to keep my viewpoints and opinions to myself because someone might disagree with me or become offended, you must be new here. I give not a fuck.

          --
          My rights don't end where your fear begins.