摘要
在等级依赖效用(RDU)框架下研究了概率扭曲对于A股市场的风险定价影响.理论上,通过推导概率扭曲下的CAPM模型,发现概率扭曲会通过影响投资者对于尾部风险的感知强弱程度而扭曲定价核,进而影响风险与收益的定价关系.当投资者低估(高估)尾部风险时,定价核呈钟型(U型),风险与收益之间负向(正向)关联.实证上,通过上证50ETF期权和A股截面数据证实了本文的理论发现.首先,由期权数据估计经验定价核,发现A股市场钟型定价核占主导,即大部分时期投资者会低估尾部风险.随后,构建概率扭曲指标将市场区分为高估时期和低估时期,回归结果发现:当市场处于低估(高估)尾部时期,系统性风险和特质风险都与预期收益显著负(正)相关.
This paper studies the effect of probability weighting on risk pricing in A-share market under the framework of rank dependent utility(RDU).Theoretically,the paper derives the RDU-CAPM and finds that probability weighting will distort the pricing kernel through distorting the perceived tail risk,and then affect the relationship between risk and return.When investors underweight(overweight)the tail risk,the pricing kernel is bell shaped(U-shaped),and the risk is negatively(positively)related to expected return.Next,by using Shanghai Stock Exchange 50 ETF option data and A-share market cross-section data,the paper confirms empirically our theoretical findings.Based on the implied pricing kernel from option data,the pricing kernel in A-share market is found to be bell shaped most of the time;that is,the investors often underweight tail risk in most times.After that,by constructing several probability weighting indices,the whole samples are divided into the over-and underweighting periods.Fama-MacBeth and Panel Regression finds that during underweighting(overweighting)periods,both systematic risk and idiosyncratic risk are significantly negatively(positively)related to expected return.
作者
石芸
芮灏
周勇
SHI Yun;RUI Hao;ZHOU Yong(School of Statistics and Academy of Statistics and Interdisciplinary Sciences,Key Laboratory of Advanced Theory and Application in Statistics and Date Science-MDE,East China Normal University,Shanghai 200062,China;School of Management,Shanghai University,Shanghai 200444,China)
出处
《管理科学学报》
CSSCI
CSCD
北大核心
2022年第10期76-95,共20页
Journal of Management Sciences in China
基金
国家自然科学基金资助项目(71971083,71931004)。
关键词
概率扭曲
等级依赖效用
系统性风险
特质波动率
probability weighting
rank dependent utility
systematic risk
idiosyncratic volatility
