Document Search
Add To My Bookshelf Sign in or Register Save And Annotate
Keywords:

spectral sign in to follow this
extreme sign in to follow this
EVT sign in to follow this
tail sign in to follow this
dependence sign in to follow this
regular sign in to follow this
variation sign in to follow this
Categories:

VaR Methods sign in to follow this
--Extreme Value Theory sign in to follow this
Half-Life:
Impact:
Discuss This Paper
Sign in to follow this page
Recent Comments
  more
Estimation of the Spectral Measure and the Tail Dependence Coefficient for Regular Varying Random Ve
Company: ETH Zentrum
Year Of Publication: 2002
Month Of Publication: October
Pages: 70
Download Count: 342
View Count:
Comment Num: 0
Language: EN
Source:
Who Can Read: Free
Date: 1-31-2007
Publisher: Administrator
Summary
The present thesis uses multivariate regular variation asa tool for modelling heavy-tailed multivariate phenomena. Extremalbehavior of such random vectors is characterized in terms of a tail indexand a spectral measure or equivalently in terms of a measure μ.An estimator of μ is proposed and proved consistent. Calibration ofthis estimator is achieved by exploiting a scaling property the measureμ possesses, an idea proposed by St˘aric˘a and extended in the presentthesis. A survey on convergence of this estimator is included. Applicationsinclude estimation of the spectral measure, conditional probabilityestimations for improbable events as well as estimation of the tail dependencecoefficient. The thesis is concluded with a case study on currencyexchange data and a brief discussion on how multivariate regular variationcompares to other approaches of multivariate extremal modelling,applicability of the model and topics of future research.
Author(s)
Wendin, Jonathan Sign in to follow this author
This document's citation network:
Similar Documents:
Close window
Sign up in one step, no personal information required. Already a Member?



Email:
Repeat Email:
User Name:
Password:
Confirm Password:

Sign Up


Welcome to GloriaMundi!
Thanks for singning up



continue or edit your profile