Open Thread

This is our monthly place to discuss relevant topics that have not appeared in recent posts.

GD Star Rating
Tagged as:
Trackback URL:
  • stephenrdiamond

    I think I recall that Robin has discussed some evidence concerning the neurophysiology of near mode and far mode, but I haven’t been able to find it. Does anyone recall where that discussion occurred or know of other research, analysis, or speculation about these underpinnings?

  • blink

    What is your take on Gordon’s recent working paper?  At least at first blush, his extrapolations seem inconsistent with your historical model of singularities and accelerating growth.

  • Evan Daniel

    Hopefully this is a stupid question, and someone can just point me to a reference. How would a conditional prediction market actually work? In the only prediction markets I’ve looked at in detail (intrade, idea futures), you buy shares, and those shares either pay out at full price or zero at market close. With a simple bet between two people, the mechanism for “calling off” the bet is simple and obvious: no money changes hands. But how does this work in a market context? You can’t simply undo all market trades, since presumably you’re interested in rewarding people who can reliably predict the market changes even when the bets get called off later.

    • Martin Epstein

      What’s wrong with refunding the money when the conditions aren’t met? You say “presumably you’re interested in rewarding people who can reliably
      predict the market changes even when the bets get called off later”. How can you possibly reward somebody for being right about, say, unemployment under the Romney administration if there is no Romney administration?

      • Peter McCluskey

         Where does there refund come from?

        If it comes from the person on the other side of the original trade, that other trader faces an unexpected loss of money from a trade he may have already taken a profit on.

        A sophisticated market will provide some way for the trader to have made another bet that offsets the risk of Romney losing.

      • Evan Daniel

         If I currently think the market value on that statistic is too low, and invest, and it rises, so I sell and exit the market, I think I should get to keep the money from that trade. It seems that, in general, it is just as legitimate to predict future market movements as it is to predict the eventual judgment.

    • Peter McCluskey

       The simple implementation is a pair of contracts such as “Obama wins and unemployment rises” and “Romney wins and unemployment rises”.

      Robin has some more sophisticated but harder to explain combinatorial market ideas, see this Daggre blog post for an example.

      • Evan Daniel

        So in the close conditional election markets, that would be “Obama wins by < X% popular vote, and unemployment rises" plus "Romney …"?

        Given a set of prices on those, how would you go about interpreting the market view of causality, as distinct from the market view as to who is more likely to win given a close election? As your definition of close gets tighter, this is less of an issue, but it also pushes your prices toward zero, and the opposing position price toward 1, which is problematic because of long-shot bias, poor effective interest rate, and lack of investor interest in highly unlikely outcomes (elections just don't get within 0.1%, for instance).

        A bet between two people that gets called off if the condition isn't met is a lot cleaner, and what I found Robin advocating in blog posts, but I'm not sure how to convert that into a market.

  • Wallace M Forman

    Anyone read “The Righteous Mind”? What did you all think?

    Towards the end of the book, Haidt attempts to explain why liberals and conservatives are different. He seems to be proposing a mostly genetic explanation – the two camps correlate with people with a disposition toward risk-seeking or risk-avoidance respectively. One extrapolation from his sparse explanation might be that the cities are filled with the descendants of the risk taking liberals who moved into those places, while the risk averse conservatives were left behind in the countryside. I find this explanation (which I have filled in) underwhelming, but I’m unsure whether Haidt really has any better explanation about why geographical regions are dominated by liberal or conservative configurations of Haidt’s “moral foundations”.

    One possible alternative (or supplementary) explanation might contrast Urban and Parochial Morality (compare “farmers and foragers”). This leans less hard on genetic differences between conservatives and liberals and harder on the differences in their environments. Humans might have an innate ability to “learn” a morality built on something like Haidt’s moral foundations (similar to an innate ability to learn a language), but not all environments would allow coordination around equivalent moralities. 

    Parochial Morality: In a small town, gossip spreads quickly, including among family, interactions are often repeated with well known friends or neighbors, and people may informally demand interaction based on enforceable norms of proportionality (I helped you, you owe me a favor; if you don’t work you don’t eat). The population in the countryside may also be much more homogeneous than the mixing pots of the city, allowing parochial morality to evoke a much clearer picture of group identity. The smaller number of people can coordinate around and enforce sacred norms (go to Church on Sundays, don’t cuss in front of the ladies, no green molds on the west side of your house). Authority matters more because you are probably in close contact with your parents, the original authorities.

    Urban Morality: Cities are dense, and filled with many anonymous strangers, so it is harder to coordinate on arbitrary sacred values (don’t eat pork, e.g.) and reputation is less important for enforcement of morality when more interactions are with anonymous strangers. Because so many different people interact in cities, a shared morality there must be much thinner – it is a lowest common denominator morality. A morality needs to be mutually acceptable to serve its purpose of tying a group together, and ultimately the only thing so many strangers can comfortably coordinate around is the unobjectionable “harm” foundation, and perhaps a fairness principle that leans less on reputation – equality, rather than proportionality.


    • stephenrdiamond

      the two camps correlate with people with a disposition toward risk-seeking or risk-avoidance respectively.

      I think of Kahneman’s frame theory and its theorem that while choosers with a good choice are risk averse, those with only bad choices are risk seeking.

  • Pingback: Overcoming Bias : Open Thread | Neuro Physiology Blog

  • William Eden

    You talk a lot about Forager-Farmer-Industrial norms, etc.

    To what degree have you looked into Pastoralist cultures, and to what degree do you think they have a different set of social rules than the other groups you routinely mention?