Using daily return data from the four major Central and Eastern European stock markets including fourteen highly liquid stocks and ATX (Vienna), PX (Prague), BUX (Budapest), and WIG20 (Warsaw) market indices, we model the value-at-risk using a set of univariate GARCH-type models. Our results show that, in both in-sample and out-of-sample value-at-risk estimations, the models based on asymmetric distribution of the error term tend to perform better or at least as well as the models based on symmetric distribution (i.e., Normal or Student) when the left tails of daily return distributions are concerned. Evaluation of the same models is less clear, however, when the right tails of the distribution of daily returns must be modelled. We suggest an asset-specific approach to selecting the correct parametric VaR model that depends not only on the risk level considered but also on the position in the underlying asset.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Publisher Info
Paper provided by Charles University Prague, Faculty of Social Sciences, Institute of Economic Studies in its series Working Papers IES with number
2008/18.