## FANDOM

33.176 Seiten

Angaben zur Quelle [Bearbeiten]

 Autor Amine Jalal, Michael Rockinger Titel Predicting tail-related risk measures: The consequences of using GARCH filters for non GARCH data Herausgeber FAME - International Center for Financial Asset Management and Engineering Datum Juni 2004 Anmerkung Research Paper N° 115 URL http://www.swissfinanceinstitute.ch/rp115.pdf Literaturverz. ja Fußnoten ja Fragmente 2

Fragmente der Quelle:
 Zuletzt bearbeitet: 2012-08-22 17:51:47 Hindemith

 Typus Verschleierung Bearbeiter Hindemith Gesichtet
Untersuchte Arbeit:
Seite: 85, Zeilen: 17-21
Quelle: Jalal Rockinger 2004
Seite(n): 1, Zeilen: 7-11
For instance, Jalal and Rockinger (2004) investigated the consequences for value at risk and expected short fall [sic] purposes of using a GARCH filter on various misspecified processes. They show that careful investigation of the adequacy of the GARCH filter is necessary since under misspecifications a GARCH filter appears to do more harm than good. We investigate the consequences for value-at-risk and expected shortfall purposes of using a GARCH filter on various mis-specified processes. We show that careful investigation of the adequacy of the GARCH filter is necessary since under mis-specifications a GARCH filter appears to do more harm than good.
 Anmerkungen Die Beschreibung der Arbeit von Jalal und Rockinger (2004) wird direkt aus dem Abstract der Publication übernommen, ohne dass dies kenntlich gemacht wäre. Sichter (Hindemith) Plagiatsfischer

 Zuletzt bearbeitet: 2012-08-22 16:46:18 Plagiatsfischer

Expression (4.5.1) is a two-tailed test that is asymptotically distributed as binomial. We perform the null hypothesis that it is a method that correctly estimates the risk measures against the alternative that the method has a systematic estimation error and gives too few or too many violations. Under the null hypothesis that a method correctly estimates the conditional quantiles, the empirical version of the statistic ${\scriptstyle \sum_{t\in T} \mathbf{1}_{\{X_{t+1} > x^t_q\} } }$ is from the binomial distribution ${\scriptstyle B (card (T), 1-q) }$. We perform a two-sided binomial test of the null hypothesis against the alternative that the method has a systematic estimation error and gives too few or too many violations.