Parameter estimation for power-law distributions by maximum likelihood methods

被引：174

作者：

Bauke, H. ^{[1
]}

机构：

[1] Univ Oxford, Rudolf Peierls Ctr Theoret Phys, Oxford OX1 3NP, England

来源：

EUROPEAN PHYSICAL JOURNAL B | 2007年 / 58卷 / 02期

关键词：

02.50.Tt Inference methods; 89.75.-k Complex systems;

D O I：

10.1140/epjb/e2007-00219-y

中图分类号：

O469 [凝聚态物理学];

学科分类号：

070205 ;

摘要：

Distributions following a power-law are an ubiquitous phenomenon. Methods for determining the exponent of a power-law tail by graphical means are often used in practice but are intrinsically unreliable. Maximum likelihood estimators for the exponent are a mathematically sound alternative to graphical methods.

引用

页码：167 / 173

页数：7

共 10 条

[1]

Bain L., 2000, INTRO PROBABILITY MA

[2]

Brent R. P., 1973, ALGORITHMS MINIMIZAT

[3]

Dress H, 2000, ANN STAT, V28, P254

[4] Problems with fitting to the power-law distribution [J].