Local Likelihood Density Estimation and Value-at-Risk. (22nd April 2010)

• Record Type: Journal Article
• Authors: Gourieroux, Christian; Jasiak, Joann
• Other Names: Zitikis Ričardas Academic Editor.
• Is Part Of: Journal of probability and statistics. Volume 2010(2010)
• Publication Date: 2010-04-22
• Subjects: Probabilities -- Periodicals; Mathematical statistics -- Periodicals; Mathematical statistics; Probabilities; Periodicals; 519
• DOI: 10.1155/2010/754851
• Languages: English
• ISSNs: 1687-952X

Abstract

Abstract : This paper presents a new nonparametric method for computing the conditional Value-at-Risk, based on a local approximation of the conditional density function in a neighborhood of a predetermined extreme value for univariate and multivariate series of portfolio returns. For illustration, the method is applied to intraday VaR estimation on portfolios of two stocks traded on the Toronto Stock Exchange. The performance of the new VaR computation method is compared to the historical simulation, variance-covariance, and J. P. Morgan methods.