A comparison of estimators for 1/f noise

被引：101

作者：

Pilgram, B

Kaplan, DT

机构：

[1] Macalester Coll, Dept Math & Comp Sci, St Paul, MN 55105 USA

[2] Graz Univ, Inst Math, A-8010 Graz, Austria

[3] McGill Univ, Ctr Nonlinear Dynam, Montreal, PQ H3G 1Y6, Canada

来源：

PHYSICA D-NONLINEAR PHENOMENA | 1998年 / 114卷 / 1-2期

基金：

加拿大自然科学与工程研究理事会;

关键词：

noise; noise parameter estimation; noise generation;

D O I：

10.1016/S0167-2789(97)00188-7

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

We use a Monte-Carlo approach to investigate the performance of five different time-series estimators of the exponent alpha in 1/f(alpha) noise. We find that a maximum-likelihood estimator is markedly superior to Fourier regression methods and Hurst exponent methods. The results indicate that useful estimates of alpha can be made from time series that are much shorter than generally presumed. (C) 1998 Elsevier Science B.V.

引用

页码：108 / 122

页数：15

共 26 条

[1]

BASSINGHTHWAIGH.JB, 1994, FRACTAL PHYSL, P78

[2] EVALUATING RESCALED RANGE ANALYSIS FOR TIME-SERIES [J].