Asymmetric Response of Volatility: Evidence from Stochastic Volatility Models and Realized Volatility

Publication Type

Conference Paper

Publication Date

10-2004

Abstract

This paper examines the asymmetric response of equity volatility to return shocks. We generalize the news impact function (NIF), originally introduced by Engle and Ng (1993) to study asymmetric volatility under the ARCH-type models, to be applicable to both stochastic volatility (SV) and ARCH-type models. Based on the generalized concept, we provide a unified framework to examine asymmetric properties of volatility. A new asymmetric volatility model, which nests both ARCH and SV models and at the same time allows for a more flexible NIF, is proposed. Empirical results based on daily index return data support the classical asymmetric SV model with a monotonically decreasing NIF. This empirical result is further reinforced by the realized volatility obtained from high frequency intraday data. We document the option pricing implications of these findings.

Keywords

Bayes factors; Leverage effect; Markov chain Monte Carlo; EGARCH; Realized volatility; Asymmetric volatility

Discipline

Applied Statistics | Econometrics

Research Areas

Econometrics

Publication

2004 Financial Econometrics Symposium

Additional URL

http://econpapers.repec.org/paper/siuwpaper/24-2004.htm

This document is currently not available here.

Share

COinS