Theoretical model to deduce a PDF with a power law tail using Extreme Physical Information
msra(2011)
摘要
The theory of Extreme Physical Information (EPI) is used to deduce a
probability density function (PDF) of a system that exhibits a power law tail.
The computed PDF is useful to study and fit several observed distributions in
complex systems. This new approach permits to describe extreme and rare events
in the tail, and also the frequent events in the distribution head. Using EPI
an information functional is constructed, and minimized using Euler-Lagrange
equations. As a solution a second order differential equation is derived. By
solving this equation a family of functions is calculated. These functions
allow describing the system in terms of a eigenstates. A dissipative term is
introduced into the model, as a relevant term to study open systems. One of the
main results is a mathematical relation between the scaling parameter of the
power law observed in the tail and the shape of the head.
更多查看译文
关键词
complex system,data analysis,power law,probability density function,open system
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要