eprintid: 1132
rev_number: 11
eprint_status: archive
userid: 36
dir: disk0/00/00/11/32
datestamp: 2012-02-20 09:03:34
lastmod: 2013-01-07 12:05:30
status_changed: 2012-02-20 09:03:34
type: monograph
metadata_visibility: no_search
creators_name: Bianchi, Alessandra
creators_name: Bregni, Stefano
creators_name: Crimaldi, Irene
creators_name: Ferrari, Marco
creators_id: 
creators_id: 
creators_id: irene.crimaldi@imtlucca.it
creators_id: 
title: Analysis of a Hurst parameter estimator based on the modified Allan variance
ispublished: submitted
subjects: HA
subjects: QA
divisions: EIC
full_text_status: none
monograph_type: technical_report
keywords: Hurst parameter, long-range dependence, self-similarity, modified Allan variance, parameter estimation, wavelets, fractional Brownian motion
note: Electronic preprint, University of Bologna, ISSN 2038-7954
abstract: In order to estimate the Hurst parameter of Internet traffic data, it has been recently proposed a log-regression estimator based on the so-called modified Allan variance(MAVAR). Simulations have shown that this estimator achieves higher accuracy and better confidence when compared with other methods of common use. Here we link it to the wavelets setting and provide an asymptotic analysis in the case the signal process is a fractional Brownian motion. In particular we show that the MAVAR log-regression estimator is consistent and asymptotically normal, providing the related confidence intervals for a suitable choice on the regression weights. Finally, we show some numerical examples.
date: 2012
date_type: submitted
id_number: 3214
institution: IMT Institute for Advanced Studies Lucca
related_url_url: http://amsacta.cib.unibo.it/3214/
citation:   Bianchi, Alessandra and Bregni, Stefano and Crimaldi, Irene and Ferrari, Marco  Analysis of a Hurst parameter estimator based on the modified Allan variance.  Technical Report       (Submitted)