.png){kind=icon}
.png){kind=icon}
Return
Nonparametric jump variation measures from options
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jeconom.2021.04.005.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper proposes a novel nonparametric method for estimating tail jump variation measures from short-dated options, which can achieve rate-efficiency and works in a general infinite jump activity setting, avoiding parametric or semiparametric assumptions for the jump measure. The method is based on expressing the measures of interest as integrals of the Laplace transforms of the jump compensator and developing methods for recovering nonparametrically the latter from the available option data. The separation of volatility from jumps is done in a novel way by making use of the second derivative of the Laplace transform of the returns, de-biased using either the value of the Laplace transform or of its second derivative evaluated at high frequencies. A Monte Carlo study shows the superiority of the newly-developed method over existing ones in empirically realistic settings. In an empirical application to S & P 500 index options, we find risk -neutral negative market tail jump variation that is on average smaller than previous estimates of it, is generated by smaller-sized jumps, and has less dependence on the level of diffusive volatility. (C) 2021 Elsevier B.V. All rights reserved.
Keywords:
Jump variation
Laplace transform
Options
Nonparametric estimation
Tail risk
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.