Discussion about applicability of prediction markets for long-term prediction!

No Gravatar

For now, prediction markets are mostly used for short- or mid-term prognosis horizons.

The pay-off for the traders is contingent upon the real event. That means they have to wait until the actual event occurs, before they know what their prognosis is worth and what the pay-off for each contract is.

What about long-term events?

For long-term prediction, this doesn’t seem to be a practicable way. Waiting till 2020 when e.g. the real unemployment rate is published does not provide an incentive to trade shares for this.

But what could be a way to construe a reasonable pay-off function?

One possible approach by Prof. Spann at the University of Passau is to run two separate prediction markets concurrently. All information about prices, traders etc. of market A would not available for traders of market B and contrary.

The pay-off for Market A would be the final price of Market B and vice versa.

So it is possible to run an virtual market for e.g. 2 months predicting the outcome of an event in 2020.Another way is to let an expert(s) assess the event an use his (their) opinions as the pay-off function.

Furthermore the own end-price of the market could be the pay-off. But then manipulation is to be expected.
I&#8217-m looking forward to discuss this challenge with you!

P.S. www.Ideosphere.com seems to be a prediction market for long term predictions, but they don’t provide a solution for the problem.

13 thoughts on “Discussion about applicability of prediction markets for long-term prediction!

  1. Robin Hanson said:

    Always remember that the relevant comparison is between different mechanisms on the *same* problem. Prediction markets may still work better than other institutions for making long term forecasts.

  2. Jason Ruspini said:

    Maybe this is something of a non-problem. Consider the familiar example of a market that one might care about in twenty years: global warming. As I’ve argued here, why not always just express your view in the “on-the-run” short term contract? You don’t have to wait until 2027. You would be levered to the present, and while your p&l may be noisy, if your long-term view is correct, you will make money. Instead of being long or short a return associated with the final integral, take your position in the “derivative”.

    If you think global temperatures will not drift linearly, but will jump in several years, then there is no position to take today anyway. If you’re unsure of the rate of change but confident of the sign, just maintain your position in the short-term contracts.

  3. James Annan said:


    I’m sure you wouldn’t claim that the FX exchange is giving remotely credible estimates of such problems as “true in 2015” and I assure you it does not on “Sea level rise 1m by 2030” either.

    As for a series of short contracts, of course betting on temp(2007)-temp(2006), temp(2008)-temp(2007)…can in principle be equivalent to a single bet on temp (2030)-temp(2005) but the distant bets still have to exist right now, so I don’t see how that solves the problem.

  4. Jason Ruspini said:

    The problem as discussed in the post is that there is no “pay off” incentive to predict an index value in 2020. Using shorter-term contracts as described provides that incentive. Tolerance to volatility is another issue.

    Discounting is something that can be relatively easily debiased, and I’m not sure in what sense the price of “true in 2015″ would be correct or not. To the extent that the FX sea level rise market is wrong, I don’t think it’s a structural problem with long-term markets as such. Not having the incentive to find information today is another issue.

  5. Robin Hanson said:

    James, FX suffers from using play money, as well as from low resources (people, time, money, etc). The most relevant way to compare two institutions is on the same problem given the same resources. You could have the greatest institution there ever was, but if you applied it to twenty second graders for an hour, it just couldn’t compete with a worse institution that had the resources of a hundred professionals for years.

  6. James Annan said:

    It doesn’t take any resources to get T2015 right, beyond those required to read the claim! What sort of extra resources do you think would solve this apparent problem with the market-based approach?

  7. Jason Ruspini said:

    I don’t know what the “right” price of T2015 should be and the contract doesn’t quite reveal time preference in the first place since buyers aren’t lending funds to sellers. Insofar as the contract can be interpreted as an interest rate, it’s pretty low at ~0.50%, though positive.

  8. Robin Hanson said:

    James, I think we can agree that T2015 would not be a problem in a well functioning real money market.

  9. Chris Masse said:

    To James Annan:

    For clarity, what is the right price of the FX Claim SLvl, for you?

    “By 2030, the greenhouse effect and other causes will have raised the average world sea level by 1 meter from its 1994 level.”


    (((I dislike sentences with three negations in them. It gives you dizziness.)))

  10. Daniel Horowitz said:

    Future truth claims can be balanced with *current* truth claims.

    e.g. “The sea level is rising”

  11. Christian Wiehe said:

    I agree with James and Jason:

    How can I solve the problem of getting NOW a prediction for e.g. the unemployment rate of the US in 2015 besides running a market till this point in time?

    I think the challenge is to find a solution, how to pay-off traders (in a satisfactory way) joining e.g. a market for 2 months from now to 4/5/2007 to predict an event in 10 years.

    What do you think about 2 parallel markets and using the outcome of the other one as the pay-off function?

    I believe that rational traders would try assess just what traders of the other market believe and don’t really integrate their own information!

    On the other hand, assuming that there is no information available about the estimation of traders from the other group, the best possible assessment of the success of the traded concepts seems to be the optimal strategy!

  12. James Annan said:

    Um…sorry for the lengthy delay.

    IMO slv1 shouldn’t be more than about $0.05, and even that depends on believing some rather speculative ideas about ice sheets for which there is precious little evidence.

Leave a Reply

Your email address will not be published. Required fields are marked *