内容简介：
多主体仿真模型是一种西方学术界新近发展起来的计算模型，它可用来模拟个体行为或个体间的互动行为。通过这种模拟方法，研究者可以探索宏观现象的微观基础。
这一方法建立在二十世纪有重大影响一些的社会学说之上，如博弈论、系统论和进化论。不同于传统的数学模型，多主体仿真模型具有或然性、突变性和整体性的特点，更接近于人类生存的社会现实。本次沙龙将简单介绍一些多主体仿真模型的案例，并讨论如何将这一方法应用于政策分析。
Agent-based models are a group of computational models, which have been developed in the western academia in recent years. An agent-based model can be used to simulate individual actions or interactions between individuals. Researchers can explore the micro foundations of macro phenomena by adopting this simulation method.
This method is based on several influential social theories in the twentieth century, such as game theory, system theory, and evolutionary theory. Different from traditional mathematical models, agent-based models have the certain properties, including probability, emergence, and wholeness, which make them more resemble human social reality. After a brief overview of several agent-based models, we will discuss the role of this new method as an enhanced policy analysis tool.
