Exact inference in diagnosing value-at-risk estimates: A Monte Carlo device

In this note a Monte Carlo approach is suggested to determine critical values for diagnostic tests of Value-at-Risk models that rely on binary random variables. Monte Carlo testing offers exact significance levels in finite samples. Conditional on exact critical values the dynamic quantile test suggested by Engle and Manganelli (2004) turns out more powerful than a recently proposed Portmanteau type test (Hurlin and Tokpavi 2006).

MoreLess

Year of publication:	2008
Authors:	Herwartz, Helmut
Publisher:	Kiel : Kiel University, Department of Economics
Subject:	Value at Risk \| Statistischer Test \| Monte-Carlo-Methode \| Theorie \| Value-at-Risk \| Monte Carlo test

More details

Series:	Economics Working Paper ; 2008-16
Type of publication:	Book / Working Paper
Type of publication (narrower categories):	Working Paper
Language:	English
Other identifiers:	582117070 [GVK] hdl:10419/27671 [Handle] RePEc:zbw:cauewp:7411 [RePEc]
Classification:	G28 - Government Policy and Regulation ; C52 - Model Evaluation and Testing ; C22 - Time-Series Models
Source:	EconStor

Persistent link: https://www.econbiz.de/10010298764