The relationships between unsystematic risk, skewness and stock returns during up and down markets

Gordon Y N Tang*, Wai Cheong Shum

*Corresponding author for this work

    Research output: Contribution to journalJournal articlepeer-review

    18 Citations (Scopus)


    A recent article published in International Business Review (12 (2003) 109) argues for the usefulness of beta as a measure of risk in international stock markets. The beta-return relationship is significantly positive (negative) when the market excess returns are positive (negative). This paper extends their study further by examining other statistical risk measures. It is well known that stock returns are non-normally distributed with significant skewness and kurtosis. Under the same conditional framework, investors are found not only compensated for bearing beta risk, but also for bearing unsystematic risk, providing evidence that international investors do not hold well-diversified portfolios. Skewness, but not kurtosis, plays a significant role in pricing international stock returns. Investors accept less positive returns for positively skewed portfolios. Total risk is significantly and positively (negatively) related to realized weekly returns during up (down) markets. Our results support previous findings and add that other statistical risk measures are also useful in explaining the cross-sectional variations in international stock returns, and hence, are relevant to portfolio managers.

    Original languageEnglish
    Pages (from-to)523-541
    Number of pages19
    JournalInternational Business Review
    Issue number5
    Publication statusPublished - Oct 2003

    Scopus Subject Areas

    • Business and International Management
    • Finance
    • Marketing

    User-Defined Keywords

    • Beta
    • Conditional CAPM
    • International markets
    • Risk-return
    • Skewness


    Dive into the research topics of 'The relationships between unsystematic risk, skewness and stock returns during up and down markets'. Together they form a unique fingerprint.

    Cite this