View by Year

Diep Duong (JOB TALK): “Volatility Predictability and Jump Asymmetry”

Diep Duong

Rutgers University, New Jersey, USA

Abstract: Many recent modelling advances in asset pricing and management are predicated on the importance of jumps, or discontinuous movements in asset returns. In particular, volatility predictability is important in numerous areas of financial econometrics ranging from the pricing of volatility-based derivative products to asset management. In light of this, a number of recent papers have addressed volatility predictability, some from the perspective of the usefulness of jumps in forecasting volatility. Key papers in this area of research include Andersen, Bollerslev, Diebold and Labys (2003), Andersen, Bollerslev and Diebold (2007), Barndoff, Kinnebrock, and Shephard (2010), Corsi (2004), Corsi, Pirino and Reno (2008), Patton and Shephard (2011), and the references cited therein. In this paper, we examine the predictive content of a variety of realized measures of jump power variations, all formed on the basis of power transformations of instantaneous returns (i.e., |r_t|^q), as first discussed in Ding, Granger and Engle (1993) and Ding and Granger (1996). More specifically, we consider jump power variations with 0 < q <= 6; and construct a variety of estimators of jump risk, including upside and downside risk, jump asymmetry (i.e., realized signed jump power variation), and truncated jump measures. Our prediction experiments use high frequency price returns constructed using S&P 500 futures data as well as stocks in the Dow 30, for the period 1993-2009 period; and our empirical implementation involves estimating members of the linear and nonlinear extended Heterogeneous Autoregressive of the Realized Volatility (HAR-RV) class of models. Our findings suggest that past "large" jump power variations help less in the prediction of future realized volatility, than past "small" jump power variations. This in turn suggests the "larger" jumps might help less in the prediction of future realized volatility than "smaller" jumps. Our empirical findings also suggest that past realized signed jump power variations, which have not previously been examined in this literature, are strongly correlated with future volatility, and that past downside jump variations matter in prediction. Finally, incorporation of downside and upside jump power variations does improve predictability, albeit to a limited extent. Overall, our findings are consistent with ABD (2007) in the concluding that continuous components dominate, when predicting volatility.

Full Text:  “Volatility Predictability and Jump Asymmetry”