Refining the asymctmetric impacts of oil price uncertainty on Chinese stock returns based on a semiparametric additive quantile regression analysis

Xie, Qichang Wu Haifeng Ma Yu

This paper provides new insights into the relationship between the crude oil volatility index (OVX) and Chinese stock returns by refining and extending the latest research of Xiao et al. (2018). Firstly, we refined the evaluation of the impact of OVX changes on stock returns by introducing nonparametric function setting, a technique robust to parametric form assumptions of Xiao et al. (2018). Secondly, we extend the original data sets from May 10, 2007, to April 9, 2020, to assess the consistency of the estimated results. Thirdly, we apply the most recent econometric model, the semiparametric additive quantile regression proposed by Fasiolo et al. (2020), instead of the linear quantile regression used in Xiao et al. (2018) to explore the potential nonlinear relationships between OVX shocks and stock returns from the asymmetric and heterogeneous perspectives. Fourthly, we investigate whether the COVID-19 pandemic mediated the OVX–stock connection, because this pandemic has had a tremendous impact on global financial markets. Our results reveal that the implications of OVX shocks on stock returns are nonlinear and dynamic and exhibit high asymmetry and heterogeneity. In particular, the positive OVX changes have a “U-shaped” effect on Chinese stock rewards in a bearish market, rather than the linear negative effect reported by Xiao et al. (2018). Moreover, the COVID-19 pandemic amplified the consequences of the positive and negative OVX variations on Chinese stock returns.