Release Time:2019-03-13  Hits:

Indexed by: Journal Article

Date of Publication: 2016-11-01

Journal: COMMUNICATIONS IN STATISTICS-THEORY AND METHODS

Included Journals: Scopus、EI、SCIE

Volume: 45

Issue: 21

Page Number: 6357-6368

ISSN: 0361-0926

Key Words: Precise large deviations; Two-dimensional; Extended regular variation distributions; Copula; Random sums

Abstract: This article focuses on the tail probabilities of the partial sums (S) over right arrow (n) = Sigma(n)(k=1) (X) over right arrow (k) and the random sums (S) over right arrow (N(t)) = Sigma(N(t))(k=1) (X) over right arrow (k), where {(X) over right arrow (k), k >= 1} is a sequence of independent identically distributed non- negative random vectorswith two dependent components ( using copulas for operational risk measurement) having extended regularly varying tails, and N( t) is a counting process independent of the sequence {(X) over right arrow (k), k >= 1}. Under some reasonable assumptions, some precise large deviation results for (S) over right arrow (n) and (S) over right arrow (N(t)) are obtained in the componentwise way.