Abstract
Our aim is to design mechanisms that motivate all agents to reveal their predictions truthfully and promptly. For myopic agents, proper scoring rules induce truthfulness. However, when agents have multiple opportunities for revealing information, and take into account long-term effects of their actions, deception and reticence may appear. Such situations have been described in the literature. No simple rules exist to distinguish between the truthful and the untruthful situations, and a determination has been done in isolated cases only. This is of relevance to prediction markets, where the market value is a common prediction, and more generally in informal public prediction forums, such as stock-market estimates by analysts. We describe three different mechanisms that are strategy-proof with non-myopic considerations, and show that one of them, a discounted market scoring rule, meets all our requirements from a mechanism in almost all prediction settings. To illustrate, we extensively analyze a prediction setting with continuous outcomes, and show how our suggested mechanism restores prompt truthfulness where incumbent mechanisms fail.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
We use the notation \(p_x\) for the density of distribution \(\varvec{p}\) at x.
- 2.
The example is true for the logarithmic scoring rule because its scores are unbounded. Every scoring rule that values exact predictions over inexact ones sufficiently will do.
- 3.
Chen et al. (2010) use the same construction to generalize from Alice-Bob-Alice to a finite-players game.
- 4.
Note that the ideal of truth-telling as dominant strategy is not attainable here, because if a player is aware of another player’s distortion, the correct Bayesian response is to compensate for the distortion.
- 5.
This formulation is different from the mechanism of prediction markets, but equivalent to it. In prediction markets, an agent replaces the current market prediction by his own. In our formulation, the agent merely announces a prediction, which, assuming the agent is truthful, becomes the market prediction by Bayesian inference. This is because all rational agents reach the same beliefs from the same data.
- 6.
Technically, an uninformative prior may be envisioned as the limit of a uniform or normal distribution as the variance goes to infinity.
- 7.
A submodularity property is required of the signal lattice, in a context described in their article.
References
Azar, Y., Ban, A., Mansour, Y.: When should an expert make a prediction? In: Proceedings of the 2016 ACM Conference on Economics and Computation, pp. 125–142. ACM (2016)
Ban, A.: Strategy-proof incentives for predictions. arXiv preprint arXiv:1805.04867 [cs.GT] (2018)
Brier, G.W.: Verification of forecasts expressed in terms of probability. Weather Rev. 78(1950), 1–3 (1950)
Chen, Y., et al.: Gaming prediction markets: equilibrium strategies with a market maker. Algorithmica 58(4), 930–969 (2010)
Chen, Y., Reeves, D.M., Pennock, D.M., Hanson, R.D., Fortnow, L., Gonen, R.: Bluffing and strategic reticence in prediction markets. In: Deng, X., Graham, F.C. (eds.) WINE 2007. LNCS, vol. 4858, pp. 70–81. Springer, Heidelberg (2007). https://doi.org/10.1007/978-3-540-77105-0_10
Chen, Y., Waggoner, B.: Informational substitutes. In: 2016 IEEE 57th Annual Symposium on Foundations of Computer Science (FOCS), pp. 239–247. IEEE (2016)
Conitzer, V.: Prediction markets, mechanism design, and cooperative game theory. In: Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence, pp. 101–108. AUAI Press (2009)
De Finetti, B.: La prevision: Ses lois logiques, ses sources subjectives. Ann. Inst. Henri Poincaré 7(1937), 1–68 (1937)
Dimitrov, S., Sami, R.: Non-myopic strategies in prediction markets. In: Proceedings of the 9th ACM conference on Electronic commerce, pp. 200–209. ACM (2008)
Gao, X.A., Zhang, J., Chen, Y.: What you jointly know determines how you act: strategic interactions in prediction markets. In: Proceedings of the Fourteenth ACM Conference on Electronic Commerce, pp. 489–506. ACM (2013)
Gneiting, T., Raftery, A.E.: Strictly proper scoring rules, prediction, and estimation. J. Amer. Statist. Assoc. 102(477), 359–378 (2007). https://doi.org/10.1198/016214506000001437. http://amstat.tandfonline.com/
Good, I.J.: Rational decisions. J. Roy. Stat. Soc. Ser. B (Methodol.) 14(1), 107–114 (1952)
Hanson, R.: Combinatorial information market design. Inf. Syst. Frontiers 5(1), 107–119 (2003)
Kong, Y., Schoenebeck, G.: Optimizing Bayesian information revelation strategy in prediction markets: the Alice Bob Alice case. In: LIPIcs-Leibniz International Proceedings in Informatics, vol. 94. Schloss Dagstuhl-Leibniz-Zentrum fuer Informatik (2018)
Tong, Y.L.: The Multivariate Normal Distribution. Springer, Heidelberg (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this paper
Cite this paper
Ban, A. (2018). Strategy-Proof Incentives for Predictions. In: Christodoulou, G., Harks, T. (eds) Web and Internet Economics. WINE 2018. Lecture Notes in Computer Science(), vol 11316. Springer, Cham. https://doi.org/10.1007/978-3-030-04612-5_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-04612-5_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-04611-8
Online ISBN: 978-3-030-04612-5
eBook Packages: Computer ScienceComputer Science (R0)