Cross-sectional return predictability: the predictive power of return asymmetry, skewness and tail risk

Xu, Zhongxiang (2017) Cross-sectional return predictability: the predictive power of return asymmetry, skewness and tail risk. PhD thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (995kB)

Abstract

This thesis attempts to investigate the cross-sectional predictive power of return asymmetry, skewness and tail risk. It mainly consists of three empirical chapters on the relation between predictive patterns of the return distribution and expected stock returns.

In the first empirical chapter, I adopt a measure of asymmetry, originally proposed by Patil et al. (2012), which can be employed to characterise the shape of the entire distribution of asset returns instead of skewness. Empirical evidence on the relation between asset returns and the skewness of the return distribution is mixed. As skewness is primarily influenced by the tail behaviour of the return distribution, it is possible for two distributions with identical skewness to have quite different asymmetry. I will examine the relationship between this new measure and stock returns. My empirical analysis indicates that stocks with high return asymmetry exhibit low expected returns. The negative relation between return asymmetry and expected returns persists after I control for size, book-to-market, momentum, short-term return reversals, liquidity, idiosyncratic volatility and various skewness factors. My results are consistent with the findings from theoretical models such as those of Brunnermeier et al. (2007) and Barberis and Huang (2008).

In the second empirical chapter, I examine the default risk and financial crisis explanations for the market skewness risk effect and find that the effect is stronger among stocks with large size, high growth, and low default risk. This suggests that the positive skewness preference theory only holds for safe stocks. Moreover, the effect of market skewness risk on stock returns interacts with default risk significantly. Market skewness risk has explanatory power for stock returns only during the periods of good economic conditions. Additionally, the market skewness risk effect is not persistent. After the financial crisis of 2007-2008, the strong effect disappears.

In the last empirical chapter, I know that investors sometimes underweight the tail event. I then try to figure out this situation by examining the default risk and financial crisis explanations for the tail risk effect. I find that market size, book-to-market ratio, and default risk have large impact on the tail risk effect. Moreover, tail risk only has explanatory power for stock returns during the periods of good economic conditions. The results suggest that when investors hold stocks with small size, low growth, and high default risk, the tail risk tends to be ignored. The tail risk effect is not persistent. The significant tail risk effect also disappear after the financial crisis of 2007-2008.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Li, Xiafei
Chevapatrakul, Thanaset
Keywords: Financial crises, Risk management, Risk-return relationships
Subjects: H Social sciences > HG Finance
Faculties/Schools: UK Campuses > Faculty of Social Sciences, Law and Education > Nottingham University Business School
Item ID: 41310
Depositing User: Xu, Zhongxiang
Date Deposited: 25 Aug 2017 10:41
Last Modified: 30 Aug 2017 08:33
URI: http://eprints.nottingham.ac.uk/id/eprint/41310

Actions (Archive Staff Only)

Edit View Edit View