Asymmetric index stock returns: evidence from the G–7

Abstract

There is able empirical evidence that the conditional volatility of stock returns is asymmetric in the sense that negative innovations increase volatility more than positive innovations of an equal magnitude. Less attention, however, has been paid to possible asymmetries in the conditional mean. This paper uses time-varying asymmetric distributions to model index stock returns of the Group of Seven (G-7) industrialized nations. In agreement with the extant literature, all index stock returns exhibit asymmetric volatility. More importantly however, the conditional mean is also an asymmetric function of past innovations, in most cases. Interestingly, the asymmetry in the conditional mean case is the reverse of that observed in the conditional variance, that is, positive innovations have a greater impact than negative innovations of an equal sign. Equivalently, positive innovations (up markets) are more persistent than negative innovations (down markets). Overall, the evidence suggests that taking into account both size as well as the sign of past innovations can improve forecasts of the conditional first and second moments of stock returns.