空间经济演化的微观机理与动态模拟

This project is going to find an effort to move models of geographical economics into the dynamic, agent-based microsimulation framework. Agent-based computational economics (ACE) works by disaggregating the modeling environment to the extent that each heterogeneous agent is represented as acting individually. Therefore, instead of making nonlinear systems tractable by modeling complex building blocks with few interactions, we can make them understandable by modeling simple building blocks with many interactions among different agents in different districts based on our analytic economic model. This disaggregation has the effect of permitting the modeled agents to act in ways that move the system out of equilibrium. ACE has a growing role in economic modeling as economists become more interested in the heterogeneity of human behavior and the effects of perception on economic outcomes. Transforming the models of geographical economics into an agent-based framework could test the assumptions and results of the economic model for robustness and find something new. This transformation also lays the groundwork for further development of models that address questions raised by geographical economics, but where assumptions of single homogeneity of actors are problematic..The project has three tasks: (1) to develop a fundamental computational platform based on ACE, ‘New’ New Economic Geography and Evolutionary Economic Geography. We plan to explore the relationships posited by basic models of geographical economics in a dynamic, agent-based microsimulation and heterogeneity framework in two city scenario and three city scenario or even more. Some initial results show that the relationships posed in the basic model do not make sense in a dynamic, micro-level model of economic decisions. Specifically, when put into a dynamic model of economic decision-making, issues such as the supply of labor, wage-setting and production levels are all mathematically impossible to resolve given the assumptions that form the foundation for the basic models. Finally, we would like to integrate the inner and intra city do some research on a multi heterogeneous situation. (2) to develop an extend computational platform by ACE focusing on urban and regional integration, technological innovation and spatial diffusion and the regional impact of carbon trading in China. (3) to develop a practical computational platform focusing on Jing-Jin-Ji integration and spatial dynamics for policy simulation..The research project is expected to contribute some innovative research to the theory and methodology of ACE in evolutionary economic geography.

吸收和借鉴国内外有关“新”新经济地理学、基于agent的计算经济学、演化经济地理学以及其他研究的最新成果，以复杂系统理论、方法和建模技术为基础，以我国国情和实际为背景，突出学科交叉、渗透和融合，将企业异质性、劳动异质性和消费者异质性纳入到统一的分析框架中，发展基于agent的计算空间经济学的理论和方法，研究进化与学习算法，实现企业生产效率、劳动技能以及消费者偏好等异质性的内生化，进而构建多重异质性下基于agent的空间经济演化模拟实验基础平台、拓展平台以及应用平台，探索两城市和多城市、城市内与城市间的企业区位选择、人口迁移、区域一体化、知识溢出、技术创新与技术扩散以及碳交易区域影响的微观基础，揭示宏观异质性与微观异质性的相互作用下经济活动的空间结构和协同演化规律，特别是多重微观异质性个体的互动及学习和适应性对宏观空间结构演化的相互影响，并以京津冀等城市群为应用案例，展开情景分析和动态模拟。

面向前沿科学领域、强调学科交叉，结合“新”新经济地理学（NNEG）、基于agent的计算经济学（ACE），开展宏观与微观异质性的相互作用下经济活动的空间结构和协同演化研究，深入比较了NEG与ACE的异同，提出了空间经济学与基于agent的计算经济学融合的理论和方法，揭示了异质性个体知识溢出的微观机理和地区生产率变化的动态规律，发现“不完全集聚”现象，发现动态迁移和知识溢出对地区生产率存在“拉高效应”与“拉低效应”，发现边缘区早期突变性创新有机会改变空间演化路径。发现不同市场开放程度下产品差异化对产业集聚存在影响，发现区域间一体化程度和变化对区域内部的一体化存在影响。超前研究了全国碳市场一体化对空间经济的动态影响和微观机理，探讨了碳配额、交易方式、区域政策等对企业区位选择、地区效率和福利等的影响，发现碳约束既不是分散力也不是聚集力。针对NEG忽略产品垂直异质性，开发了FC-Q模型和FE-Q，探讨了产业集聚对地区产品质量的影响和动态，发现在同质性假设下产业集聚未必提高产品质量，但区域一体化和地方品质提高有助于产品质量提升，产品质量是一种分散力。引入企业和消费多重异质性，讨论了产业集聚、企业效率与产品质量的互动关系，发现质效双重空间选择效应，且地区产品质量的高低取决于要素是否流动、需求偏好程度以及市场规模。针对京津冀协同演化与非首都功能疏解展开计算实验，发现疏解效果取决于周边地区的条件，模拟了北京医疗服务设施疏解与空间布局优化，比较了“点轴模式”与“组团模式”，发现组团模式在应急状态下要优于点轴布局。利用CGE分析非首都功能疏解对北京产业结构的影响。开发SD模型讨论新冠疫情对北京经济的冲击。建立了基于空间经济学理论的国土空间演化模拟分析平台，强调集聚经济是国土空间演化格局的重要驱动力，修正了“双评价”，提高了科学性。围绕经济集聚、环境规制、绿色TFP、产业网络及技术关联等展开多维度、多视角和多尺度的特征事实分析、实证检验和案例研究。