Information Transmission and Aggregation

信息传输与聚合

• 批准号: 1155931
• 负责人: Navin Kartik
• 金额: $ 28.72万
• 依托单位: Columbia University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2012
• 资助国家: 美国
• 起止时间: 2012-05-01 至 2018-04-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1155931&HistoricalAwards=false
• 关键词: Information; Transmission; Aggregation

This award funds research that uses game theory to analyze situations where information has an important effect on how individual decisions lead to group outcomes.The first part of the project considers whether and how political systems lead to the implementation of efficient policies. A key feature of any real world political system is that information is dispersed across citizens, policy-makers, government executives, and legislators. Social scientists now have a fairly well developed understanding of how elections aggregate voters' information in a representative democracy, but we do not yet know whether or not politicans' or parties' information is also disseminated and aggregated in an efficient way. The researcher will use a novel model of electoral competition between office-motivated candidates with private information to answer this question. The second part of the research develops a new approach to studying strategic information transmission. The key idea is that conventions of natural language can be formally captured by imposing a weak restriction on the space of available strategies. While the restriction by itself does nothing to affect the essence of the strategic possibilities, it mitigates one source of multiplicity in games. Given the formal restruction capturing conventions, the second step is to apply iterative admissability, ie to iteratively delete weakly dominated strategies. The new approach provides sharp and robust predictions. The project includes an investigation of this approach to broader classes of games.The projects here will be valuable for many applications of game theory modelling. The project also contributes to training graduate students in how to use game theory to model concrete situations.

这个奖项资助了一项研究，该研究使用博弈论来分析信息对个人决策如何导致集体结果具有重要影响的情况。该项目的第一部分考虑政治制度是否以及如何导致有效政策的实施。任何现实世界政治制度的一个关键特征是，信息分散在公民、政策制定者、政府高管和立法者之间。社会科学家现在对选举如何在代议制民主制度下汇总选民的信息有了相当深入的了解，但我们还不知道政客或政党的信息是否也以有效的方式传播和汇总。研究人员将使用一种新的选举竞争模型来回答这个问题，这些竞争是出于公职动机的候选人之间的私人信息。研究的第二部分为研究战略信息传播开辟了一条新的途径。其关键思想是，通过对可用策略的空间施加弱限制，可以正式捕获自然语言的约定。虽然限制本身并不会影响战略可能性的本质，但它减轻了游戏中多样性的一个来源。给定形式的重构捕获约定，第二步是应用迭代可接受性，即迭代地删除弱支配策略。新的方法提供了准确而有力的预测。该项目包括对这种方法在更广泛的游戏类别中的调查。这里的项目将对博弈论建模的许多应用程序具有价值。该项目还有助于培训研究生如何使用博弈论对具体情况进行建模。