EconStor >
Rheinische Friedrich-Wilhelms-Universität Bonn >
Bonn Graduate School of Economics (BGSE), Universität Bonn >
Bonn Econ Discussion Papers, Bonn Graduate School of Economics (BGSE), Universität Bonn >

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10419/22883
  
Title:Modeling and Forecasting DAX Index Volatility PDF Logo
Authors:Lazarov, Zdravetz
Issue Date:2004
Series/Report no.:Bonn econ discussion papers 2004,5
Abstract:The recent introduction of the realized variance measure defined as the sum of the squared intra-daily returns stamped on some high frequency basis has spurred the research in the field of volatility mod- eling and forecasting into new directions. First, the realized variance is a much better estimate of the latent volatility than the sum of the weighted daily squared returns. As such it is better suited for com- paring the out-of-sample performances of competing volatility models. Additionally, it can enter as a parameter in these models proving bet- ter information than the daily returns commonly used in the standard volatility models. These two innovations have been utilized in several recent papers. We extend this line of research by estimating and com- paring a wide class of volatility models for the DAX index futures that use the realized variance or the daily returns. To give a new view of the question whether time series volatility models or implied volatility have better predictive power we estimate a model which incorporates both the historical realized variance and the historical implied volatil- ity. Our results suggest that using realized variance leads to superior performance compared to the previous approaches. Also, the inclusion of the implied volatility produces a slight improvement.
Subjects:Forecasting
High-Frequency Data
Volatility
JEL:C22
G10
Document Type:Working Paper
Appears in Collections:Bonn Econ Discussion Papers, Bonn Graduate School of Economics (BGSE), Universität Bonn

Files in This Item:
File Description SizeFormat
bgse5_2004.pdf443.23 kBAdobe PDF
No. of Downloads: Counter Stats
Download bibliographical data as: BibTeX
Share on:http://hdl.handle.net/10419/22883

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