Higher moments and US industry returns: realized skewness and kurtosis

Purpose The purpose of this paper is to examine the relationships between higher moments returns (realized skewness and kurtosis) subsequent at industry level, with a focus on both empirical predictability practical application via trading strategies. Design/methodology/approach Daily for 48 US industries over period 1970–2019 from Kenneth French’s data library are used calculate construct short- medium-term single-sort analysis adjusts common risk factors (market, size, value, investment, profitability illiquidity) confirm whether conventional asset pricing models can capture these relationships. Findings Past positively relates relationship unexplained by factors. There also time-varying effect in which predictive role much stronger business cycle expansions than recessions, result consistent varying investor optimism. However, there no significant kurtosis returns. confirms robustness using value- equal-weighted Research limitations/implications calculation realized conventionally uses high-frequency intra-day data, regrettably unavailable industries. In addition, chosen portfolio-sorting method may omit some information, as it compares only average group Nonetheless, close future level suggests variations This enriches knowledge market anomalies questions yet again weak-form efficiency validity models. One suggestion that possible significantly improve existing multi-factor including factor. Practical implications Given investors predict select better-performing funds. They even strategies based return distributions would generate abnormal Further, evaluation individual stocks contains better performance earn returns, risks related shed light stock picking. Originality/value While abundant evidence firm little level. testing time variation yields novel concerning asymmetric cycles. Finally, supplements firm-level results focusing decomposed components moments.