互联网加剧投资者有限理性研究综述

How Does Internet Escalate Investor Bounded Rationality?A Literature Review

通过对已有研究的梳理和述评,本文总结了投资者有限理性的理论内涵与行为表征,并重点从媒介信息效应视角,阐释了互联网时代媒介信息诱发或者加剧投资者有限理性行为、影响投资者决策及资产价格的基本方式;进一步从决策感染视角,解析了投资者有限理性行为借助于互联网时代日益增强的社会化交互功能的传导、演化与嬗变机制。本文意在呼吁未来的研究融合不同视角,引入跨学科方法,从行为金融的实证和建模研究中提取投资者有限理性决策规则,从互联网媒介信息中挖掘冲量效应变量,提取投资者交互网络构成、投资者信念与偏好状态等动态特征,将市场指标作为反馈,构建动力学模型,以更有效地揭示有限理性投资者决策感染以及资产价格的动态涌现机制。

Prior investigations from different disciplines probe into three aspects by following their own diagrams in an isolated way, resulting in obvious gaps and hedges. The empirical research regarding investor bounded rationality sheds substantial light on the measurement, effects, prediction and modeling of two prominent components of bounded rationality, i.e., attention and sentiment. However, the empirical evidence is not effectively consolidated with psychological descriptive frameworks and economic modeling frameworks, and is insufficient in reaching an effective model of asset pricing driven by bounded rationality, and inadequate in picturing the bounded rationality of investors. The literature on media information identifies different effects, such as the effects of category, sentiment, quantity, timing, salience, and co-movement, etc., establishing that media information significantly affects investor decision-making and financial markets, by allocating investor attention and evoking investor sentiments. Internet-channeled media information provides important input data of reality for mining for signals of asset prices, and serves as an important nexus between the decision- making of individual investors and the evolution at the societal level. Concerning behavior contagion, some of the literature empirically reveals the peer effects in financial markets, while the majority of research, enlightened by physical models, attempts to explain the anomalies in financial markets by simulating the contagion and evolution of investor decision and psychology at the system level driven by investor interactions. However, these system-level models always simplify the processes of investor decision-making and interaction, lacking a precise and detailed portray of heterogeneous investors at the individual level and lacking the input of reality data. In response to the above isolations, the current research seeks to review recent literature from the perspectives of media effects and behavioral contagion to obtain insights for understanding how media information and social interaction in Internet age arouse or even escalate the bounded rationality of investors. Particularly, by the scrutinizing, this research appeals to future scholarly efforts to synthesize the reviewed perspectives into a holistic dynamic model in an interdisciplinary way. In the synthesized model, individual decision-making rules can be inferred from the mass of prior empirical and modeling investigations on investor bounded rationality; the momentum of media, the structure of investor interaction network, as well as the dynamic states of investor belief and preference, can be learned from the real-world media information by approaches of text-mining, information extraction and sentiment analysis; and moreover, the market indicators can serve as the feedback of reality. In this synthesizing way, the dynamic emergent effects of the behavioral contagion of investors with bounded rationality and the asset prices can be more effectively and deeply unraveled.