The Aumann-Serrano risk factor and asset pricing: evidence from the Chinese A-share market

Abstract

Empirical evidence shows that a single-factor model using the Aumann-Serrano riskiness index dominates both the CAPM and Fama-French three-factor model because the index captures information on higher-order moments

Acknowledgments

The authors acknowledge help from our research assistants, Ms Shuwan Wang and Ms Jingyu Zhu, from the School of Finance, Renmin University of China.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

† The beta coefficient for each asset is calculated as the excess return of the asset over per unit of market excess return in the classic CAPM.

‡ For details of the reform and its implications, see Jiang et al. (2008), Ahn and Cogman (2007) and Hung et al. (2015). For more recent development in the Chinese stock market, please resort to a review by Liu and Wang (2017).

† Homm and Pigorsch (2012) give a brief review on the NIG distribution.

‡ Homm and Pigorsch (2012) also provides a non-parametric estimator for the AS index. Our baseline results are robust if the AS index is calculated with the non-parametric estimator. These further results are available upon request.

† Note that although the adjusted betas preclude the effect of real option values, they are still constructed based on the CAPM beta of firm i. As we have discussed, it does not satisfy the duality axiom and hence may cause potential measurement errors as a risk indicator.

‡ One simple example for this argument is the US stock market return during the financial crisis starting from mid-2007. The values of SPX returns based on a 20-day rolling window in November of 2007 are all negative.

† As opposed to the A-share stocks, there is also a B-share stock market in mainland China (excluding Hong Kong), which is denominated in RMB but shares can only be traded using US dollar or Hong Kong dollar. Besides the A- and B-share stocks, there is also a H-share stocks, which refers to the shares of companies incorporated in mainland China but are traded in the Hong Kong Stock Exchange.

‡ In the Chinese stock market, a reversal effect prevails rather than the momentum, and that momentum strategies are profitable only at short time horizons (within 4-week holding periods). For a survey in this field, please see Pan and Xu (2011). It is possible that this is caused by an overreaction of market participants (Gang, Qian and Xu, 2018).

§ For the NIG distribution and its application, please see Andersson, Bollerslev et al., Eriksson et al., Zakamouline and Koekebakker, Homm and Pigorsch (2012).

Additional information

Funding

This work was supported by National Natural Science Foundation of China [grant numbers 71503257 and 71773126].