非线性资产定价模型与特质波动率之谜  

Nonlinear asset pricing model and the idiosyncratic volatility puzzle

作　　者：邢红卫 刘维奇 Xing Hongwei;Liu Weiqi(Institution of Management and Decision, Shanxi University, Taiyuan 030006, China;Faculty of Finance and Banking, Shanxi University of Finance and Economics, Taiyuan 030006, China)

机构地区：[1]山西大学管理与决策研究所,山西太原030006 [2]山西财经大学财政金融学院,山西太原030006

出　　处：《系统工程学报》2019年第3期357-371,共15页Journal of Systems Engineering

基　　金：国家自然科学基金资助项目(71371113);教育部人文社会科学研究青年基金资助项目(16YJC790113;18YJC630-175);山西省高等学校哲学社会科学资助项目(2017203)

摘　　要：在资产收益分布非正态性或者投资者效用函数比二次型更复杂的条件下,依据均衡定价理论推导出一个基于消费的非线性资产定价模型.该模型不仅捕捉了收益的波动风险,而且可以捕捉收益的高阶矩风险.通过实证检验,发现包含因子平方项的非线性资产定价模型比传统的线性资产定价模型有更好的定价效率.分别以非线性定价模型和传统线性定价模型对"特质波动率之谜"进行检验,发现基于非线性定价模型估计特质波动率会使"特质波动率之谜"明显减弱,说明非线性资产定价模型比线性资产定价模型能够更有效地解释这一典型的市场异象.Assuming that asset returns do not obey the normal distribution and the investor’s utility function is more complex than the quadratic form, a consumption-based nonlinear asset pricing model is proposed according to the equilibrium pricing theory. The model not only captures the volatility risk of asset returns, but also captures the high order moment risk of asset returns. Empirical tests find that compared with traditional linear asset pricing models, the nonlinear asset pricing model including quadratic terms of those factors has better pricing efficiency. The idiosyncratic volatility puzzle is examined using the nonlinear and traditional linear asset pricing models, respectively. It is found that the idiosyncratic volatility puzzle decreases significantly when estimating the idiosyncratic volatility with the nonlinear asset pricing model, which indicates that the nonlinear asset pricing model can more effectively explain the typical stock market anomaly than the linear asset pricing model.

关 键 词：特质波动率 非线性模型 定价效率 错误定价 

分 类 号：F830.9[经济管理—金融学]