摘要：This note makes two contributions by extending the analysis in Bali and Peng (2006) which investigates the risk-return tradeoff at the daily horizon using high-frequency data. Our first contribution is to show that the empirical relation between returns and risk is not validated for recent years. Our second contribution is to assess the importance of disentangling jumps from the continuous component using high-frequency data and recent nonparametric methods. We show that similar results are obtained using either realized variance or an alternative measure of realized variance which is robust to jumps thereby providing evidence that jumps do not improve significantly the explanatory power in the risk-return relation.
