讲座将介绍一个新的EITM(理论模型的实证意义)来研究联盟政治和制度制衡,重点关注大国如何在联合国大会的竞争战略环境中使用经济政策工具来影响弱国的投票行为。该方法是对Signorino和他的合作者研究(1999,2003,2003)的扩展。它建立了一个三个参与人贝叶斯动态博弈模型来模拟包含两个发送者和一个共同目标国竞争联盟的形成。然后将理论模型转化为贝叶斯策略统计估计值,以克服递归和多边相互依赖带来的估计挑战。根据2003-2014年111个国家的数据,本人调查了联合国大会上投票联盟的偏好、相互依赖策略和投票联盟分界。该研究提出的EITM方法显著改进了现有的战略统计估计,且在国际关系竞争性联盟的形成研究中,具有广泛的应用潜力。
In this lecture, I will introduce a novel EITM (Empirical Implications of Theoretical Model) to study coalitional politics and institutional balancing, focusing on how great powers use economic policy instruments to maneuver voting behavior of weaker states in a competitive strategic environment of the United Nations General Assembly. This approach is an extension to Signorino and coauthors (1999, 2003, 2003), and it builds a three-player Bayesian dynamic game to model competitive coalition-formation involving two senders and one common target. Then the theoretical model is converted into a Bayesian strategic statistical estimator to overcome estimation challenges caused by recursive and multilateral interdependence. Based on data of 111 countries in 2003-2014, I investigate preferences, interdependent tactics, and division lines of voting coalitions in UNGA. The proposed EITM approach significantly improves the extant strategic statistical estimators, and has a broad potential of applications to competitive coalition formation in international relations.