摘要
本文构建了理性预期均衡模型,假定交易者过度自信,认为自己拥有最独特的信息,高估对手信息相关系数,且对信息相关系数存在暧昧.在过度自信的情绪下,交易者因低估对手信号反映的信息量而减少从价格中学习他人信号,这缓解了搭便车的行为,使交易量增加.引入信息相关性暧昧环境后,随着风险资产价格从高到低,交易者的情绪表现为盲目自信(价格高位)-观望-谨慎乐观-观望-盲目自信(价格低位)五个阶段.我们的研究发现:虽然过度自信会放大风险资产收益波动率,但是当市场充分竞争时,其增加股权溢价的作用占主导,能提升社会福利.特别是在熊市末期和反转初期,交易者表现出盲目自信,这种情绪有助于社会福利水平恢复.进一步地,交易者盲目自信时,信号暧昧程度越强,过度自信越高,投资增加,能提升整体社会福利;但当交易者谨慎乐观时,信号暧昧性的作用相反.
In this paper we adopt a rational-expectations equilibrium model,supposing traders are overconfident,believe they obtain the most unique message,and overestimate counterparties’information correlation.Meanwhile,information correlation is ambiguous.Under such emotion,traders are less likely to learn from price as they underestimate the amount of information reflected in their opponents’signals,which mitigates free-riding behavior and increases trading volume.Under ambiguous environment and with risk asset price rising,traders’sentiment develops in five stages:Blind confident(with price at the top),wait and see,cautiously optimistic,wait and see,and blind confident(with price at the bottom).It is shown that,though overconfidence intensifies return’s excessive volatility,it plays a dominant role in enhancing equity premium when the market is fully competitive,and thus improving social welfare.It is particularly effective at the end of a bear market and the beginning of a reversal when traders reveal blind confident.Besides,overconfidence degree affects the price momentum,but has no impact on the market efficiency.Finally,when traders are blind confident,information correlation ambiguity increases their confidence level,excessive investment and social welfare;when traders are cautiously optimistic,it has an opposite effect.
作者
邓晴元
刘舟
张顺明
DENG Qingyuan;LIU Zhou;ZHANG Shunming(School of Finance,Renmin University of China,Beijing 100872,China)
出处
《系统工程理论与实践》
EI
CSSCI
CSCD
北大核心
2022年第7期1755-1769,共15页
Systems Engineering-Theory & Practice
基金
国家自然科学基金(71773123,72173125)
国家社会科学基金重点项目(21AZD028)。
关键词
过度自信
相关性暖昧
理性预期均衡
信号学习
社会福利
overconfidence
correlation ambiguity
rational-expectations equilibrium
signal learning
social welfare
