Information elicitation and sequential mechanisms.

*(English)*Zbl 1281.91113Authors’ abstract: We study an incomplete information mechanism design problem with three peculiarities. First, access to agents’ private information is costly and unobservable. Second, the mechanism may communicate sequentially with the agents. Third, the mechanism designer and all the agents share a common interest. As an example one can think of \(N\) geologists that study the potential oil reserves in some tract. The geologists agree on the right course of action, given their \(N\) studies. However, carrying out the study may be costly for a geologist and so he may opt to fabricate a study. The oil company that employs these geologists need not contract them simultaneously and may, furthermore, choose to provide some of the results of early studies to geologists employed later on. Finally, the geologists and the oil company would like the joint study to forecast the quantity of oil reserves as accurate as possible. It turns out that, in such settings, what may not be implementable without communication becomes implementable with communication. Clearly, the possibility for sequential communication introduces a lot of complexity to the design problem. However, we provide a result in the spirit of the revelation principle and argue that whenever implementation is possible with communication it is also possible with a simple communication mechanism. Formally, we extend the model and results from the paper [R. Smorodinsky and M. Tennenholtz, Games Econ. Behav. 55, No. 2, 385–406 (2006; Zbl 1100.94021)] the authors of which consider a similar problem but restrict attention to symmetric social choice functions and i.i.d. distributions over the private information.

Reviewer: Vitaly Pruzhansky (Brussels)

##### MSC:

91B44 | Economics of information |

91A80 | Applications of game theory |

91A10 | Noncooperative games |

91B06 | Decision theory |

PDF
BibTeX
XML
Cite

\textit{I. Aricha} and \textit{R. Smorodinsky}, Int. J. Game Theory 42, No. 4, 931--946 (2013; Zbl 1281.91113)

Full Text:
DOI

##### References:

[1] | Battaglini, M, Sequential voting with abstention, Games Econ Behav, 51, 445-463, (2005) · Zbl 1099.91037 |

[2] | Dekel E, Piccione M (1997) On the equivalence of simultaneous and sequential binary elections. Discussion Paper No. 1206, Northwestern University, Center for Mathematical Studies in Economics and Management Science · Zbl 1079.68037 |

[3] | Gershkov, A; Szentes, B, Optimal voting scheme with costly information acquisition, J Econ Theory, 144, 36-68, (2009) · Zbl 1154.91378 |

[4] | McGrew R, Porter R, Shoham Y (2003) Towards a general theory of non-cooperative computation. In: Proceedings of the 9th conference of theoretical aspects of rationality and knowlede (TARK 2003), pp 59-71 · Zbl 1106.91026 |

[5] | Myerson, R, Optimal coordination mechanisms in generalized principal-agent problems, J Math Econ, 10, 67-81, (1982) · Zbl 0481.90001 |

[6] | Myerson, R, Mechanism design by an informed principal, Econometrica, 51, 1767-1797, (1983) · Zbl 0521.90104 |

[7] | Myerson, R, Games with communication, Econometrica, 54, 328-358, (1986) · Zbl 0599.90134 |

[8] | Persico, N, Committee design with endogenous information, Rev Econ Stud, 71, 165-192, (2004) · Zbl 1106.91026 |

[9] | Ronen A, Wahrmann L (2005) Prediction games. In: The 1st workshop on internet and network economics (WINE) · Zbl 1099.91037 |

[10] | Shoham, Y; Tennenholtz, M, Non-cooperative computation: Boolean functions with correctness and exclusivity, J Theor Comput Sci, 343, 97-113, (2002) · Zbl 1079.68037 |

[11] | Smorodinsky, R; Tennenholtz, M, Overcoming free riding in multi-party computationsâ€”the anonymous case, Games Econ Behav, 55, 385-406, (2006) · Zbl 1100.94021 |

This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.