基于网络理论的金融传染与投资者行为研究

From the new perspective of network and utilizing the leading methods of social and economic networks and complex networks, this project analyzes the accumulation and transmission of shocks in the financial system from the view of macro aspect, and studies investors’ behavior contagion, social learning and social interacting as well as the implications for asset price in the micro level. (1) The study estimates the probability of financial contagion and magnitude of expected losses when the joint distribution of shocks is given. Moreover, it is recomputed under the different financial network topology and the localized crisis of confidence. (2) The study establishes the social network interactive model of institutional investors based on the education background, professional experience and heavily held stocks of institutional investor network, investigating the influence of interactive behavior on investment decision and asset price. (3) The study constructs an individual investor behavior contagion model in multiplex networks, and empirically examines how the information spreads in the structures of multiplex networks through the investor behavior data from Sina Weibo, Weixin and Xueqiu network. (4) The study also conducts the empirical test on the relationship between investor consensus expectation and factors such as asset prices, social learning speed and network structure on the basis of professional investor behavior data from Xueqiu. This project will contribute to the research on financial contagion, cognitive patterns, decision-making and group behavior and the comprehension of asset price forming.

本项目从网络这一新视角，融入社会经济网络和复杂网络领域的前沿方法，宏观方面分析外部冲击在金融体系中的传导和演变机制，微观方面研究投资者行为传染、社会学习、社会互动及其对资产价格的影响。（1）基于外部冲击的概率分布分析金融机构的传染概率以及资产期望损失，并研究不同金融体系网络结构和市场信心对其演化的影响。（2）在建立以教育背景、供职经历、重仓股票为“链接”的机构投资者网络的基础上，构建机构投资者社会网络互动模型，研究行为互动与投资决策、资产价格的关系。（3）构建多层网络结构下的个人投资者行为传染模型，并利用新浪微博、微信、雪球网投资者行为数据，实证检验信息如何在多层网络中扩散。（4）基于雪球网投资者行为数据，实证检验投资者一致预期与资产价格、社会学习速度与网络结构的关系。本项目从宏观和微观两个方面将推进对金融危机传染、投资者认知规律、决策过程和群体行为的研究，深化对资产价格形成的认识。

随着网络技术的发展，金融中的个人和机构投资者以及他们投资的标的都成为了社会网络和经济网络中的节点。研究通过网络结构中信息和情绪的传播和传染，个体和机构投资者的投资行为对资产价格产生的影响，将推进对金融危机传染、投资者认知规律、决策过程和群体行为的研究，深化对资产价格形成的认识。本项目从宏观方面分析外部冲击在金融体系中的传导和演变机制，微观方面研究投资者行为传染、社会学习、社会互动及其对资产价格的影响。本项目研究发现，多层次多结构的社交网络结构会影响信息的传播速度与强度，并通过基金的社会和持股网络，对市场的流动性和波动性形成系统性的影响。本项目的关键数据为社交网络中的信息传导数据，基金的社会网络数据和以基金重仓股为节点的持股网络数据。本项目的主要结论将为相关金融管理者，提供通过社交网络和机构社会经济网络，进行系统风险和危机防范的数据支持。