Presentation 2003/11/14
Comparison of Estimation Methods in Traffic Data Analysis Based on Extreme Value Theory : Comparison of Maximum Likelihoood Method and Matching Moment Method
Masato UCHIDA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) The behavior of the tail distributions of the variables concerning teletraffic states, such as packet loss rate, queue length, delay time, link-usage rate, and throughput, greatly influences the communication quality, application performance, and network design, etc. Against this background, this paper uses Extreme Value Theory (EVT) to analyze traffic-data and compares two estimation methods, Maximum Likelihood Method (MLM) and Matching Moment Method (MMM), that are needed for the analysis. Though MLM is used in various fields of data analysis because it has asymptotic efficiency and asymptotic consistency, it does not have gross-error-robustness. So, it is important to compare the estimated results derived from MLM and MMM. In this paper, we disucuss their characteristics.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) P2P / Gnutella Protocol / Degree Distribution
Paper # NS2003-186,CQ2003-103,TM2003-64
Date of Issue

Conference Information
Committee NS
Conference Date 2003/11/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Network Systems(NS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Comparison of Estimation Methods in Traffic Data Analysis Based on Extreme Value Theory : Comparison of Maximum Likelihoood Method and Matching Moment Method
Sub Title (in English)
Keyword(1) P2P
Keyword(2) Gnutella Protocol
Keyword(3) Degree Distribution
1st Author's Name Masato UCHIDA
1st Author's Affiliation NTT Service Integration Laboratories, NTT Corporation()
Date 2003/11/14
Paper # NS2003-186,CQ2003-103,TM2003-64
Volume (vol) vol.103
Number (no) 443
Page pp.pp.-
#Pages 6
Date of Issue