Extreme Value Analysis of Teletraffic Data

Computational Statistics and Data Analysis (Special Issue on Computer Security and Statistics), Vol. 45, pp. 85-103, 2004

19 Pages Posted: 26 Nov 2006

See all articles by Zoi Tsourti

Zoi Tsourti

Athens University of Economics and Business

John Panaretos

Athens University of Economics and Business

Abstract

An empirically verified characteristic of the expanding area of Internet is the longtailness of phenomena such as cpu time to complete a job, call holding times, files lengths requested, inter-arrival times and so on. Extreme values of the above quantities are liable to cause problems to the efficient operation of the network and call for effective design and management. Extreme-value analysis is an area of statistical analysis particularly concerned with the systematic study of extremes, providing useful insight to fields where extreme values are probable to occur and have detrimental effects, as is the case of teletraffics. In this paper we illustrate the main elements of this analysis and proceed to a detailed application of extreme-value analysis concepts to a specific teletraffic data set. This analysis verifies, too, the existence of long tails in the data.

Keywords: Teletraffic engineering, Long tails, Extreme-value index, Smoothing procedures

Suggested Citation

Tsourti, Zoi and Panaretos, John, Extreme Value Analysis of Teletraffic Data. Computational Statistics and Data Analysis (Special Issue on Computer Security and Statistics), Vol. 45, pp. 85-103, 2004, Available at SSRN: https://ssrn.com/abstract=947061

Zoi Tsourti

Athens University of Economics and Business ( email )

76 Patission Street
Athens, 104 34
Greece

John Panaretos (Contact Author)

Athens University of Economics and Business ( email )

29 Ithakis Str.
Athens, 11257
Greece

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
93
Abstract Views
792
Rank
523,135
PlumX Metrics