Value-at-risk on Central and Eastern European stock markets: An empirical investigation using GARCH models

Arbeitspapier

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.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: IES Working Paper ; No. 18/2008

Classification and Topics

• Classification: Wirtschaft ; Semiparametric and Nonparametric Methods: General ; Multiple or Simultaneous Equation Models: Time-Series Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models; Diffusion Processes; State Space Models ; Model Evaluation, Validation, and Selection ; Forecasting Models; Simulation Methods ; Asset Pricing; Trading Volume; Bond Interest Rates
• Subject: Value-at-Risk ; Expected Shortfall ; Backtesting ; Aktienmarkt ; Risikomaß ; ARCH-Modell ; Schätzung ; Osteuropa

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Bubák, Vít
• Event: Veröffentlichung
• (who): Charles University in Prague, Institute of Economic Studies (IES)
• (where): Prague
• (when): 2008

Further information

• Handle: http://hdl.handle.net/10419/83344
• Last update: 10.03.2025, 11:44 AM CET