Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/25335 
Year of Publication: 
2009
Series/Report no.: 
SFB 649 Discussion Paper No. 2009,019
Publisher: 
Humboldt University of Berlin, Collaborative Research Center 649 - Economic Risk, Berlin
Abstract: 
As a function of strike and time to maturity the implied volatility estimation is a challenging task in financial econometrics. Dynamic Semiparametric Factor Models (DSFM) are a model class that allows for the estimation of the implied volatility surface (IVS) in a dynamic context, employing semiparametric factor functions and time-varying loadings. Because financial asset volatilities move over time, across assets and over markets, this paper analyses volatility interaction between German and Korean stock markets. As proxy for the volatility, factor loadings series derived from a DSFM application on option prices are employed. We examine volatility transmission between the markets under the vector autoregressive (VAR) model framework. Our results show that a shock in the volatility of one market may not translate directly into greater uncertainty in another market and it is unlikely that portfolio investors can benefit from diversification among these markets due to cointegration.
Subjects: 
implied volatility surface
dynamic semiparametric factor model
VAR
cointegration
JEL: 
C14
G12
Document Type: 
Working Paper

Files in This Item:
File
Size
894.25 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.