Találati lista | Tudóstér

001-es BibID:	BIBFORM072716
Első szerző:	Zapodeanu, Daniela
Cím:	Backtesting value at risk models in the presence of structural breaks on the Romanian and Hungarian stock / Zapodeanu Daniela, Kulcsár Edina, Cociuba Mihail Ioan
Dátum:	2014
ISSN:	1222-569X 1582-5450
Megjegyzések:	Transactions on financial markets are associated with variability, risk anduncertainty, so quantification of risk has a great importance. Beside Standard Deviationand Variance, one of the most involved risk measure methods is Value-at-Risk (VaR). Inthis study, we use daily return for the stock index from Romania (BET) and Hungary(BUX) for the 01:2007 - 02:2013 periods in order to test the influence of structural breakson the VaR metrics. We find out that the ARCH phenomenon is present, so we use theGARCH family models. The structural breaks in the series mean and variance areidentified using the Zivot-Andrews test and PELT algorithm, the structural break datesare captured using dummy variables in the GARCH models (struc-GARCH), the selectionof models is done using the informational criterion [Akaike, Schwarz, Log-likelihood]. Theresults of present research show a greater volatility associated with a higher risk level incase of Romanian stock index. The stock market indices return follows a negativelyskewed and leptokurtic distributions forms either in two cases, so is unspecific a normaldistribution. After applying above mentioned tests we can conclude that there are eightstructural breaks in BET index returns variance and there are five breakpoints in case ofBUX. The breakpoints in mean show very closely results in time, for BET in February2009 and for BUX March 2009. Backtesting VaR models are done by measuring thenumber of times the loss is greater than the VaR forecast. The first step for unconditionalcoverage testing consists in comparing of fraction of VaR violation for a particular riskmodel. The independence testing it is very important tool in back-testing, because it isnot the same that the VaR violations are differentiated in time or there are clustered insome certain period. By checking the independence test, we have the possibility todiscover and reject the model with clustered hit sequence. Testing the influence ofstructural breaks on VaR we find that incorporating structural breaks in the GJR-GARCHmodels generates lower violations when comparing with the plain GJR-GARCH orRiskMetrics methodology.
Tárgyszavak:	Társadalomtudományok Közgazdaságtudományok idegen nyelvű folyóiratközlemény külföldi lapban stock market structural break Value at Risk GARCH
Megjelenés:	Analele Universităţii din Oradea. Ştiinţe economice = Annals of University of Oradea. Economic science 1 (2014), p. 802-812. -
További szerzők:	Kulcsár Edina (1987-) (közgazdász) Cociuba Mihail Ioan
Internet cím:	Szerző által megadott URL Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon: