We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Physics > Chemical Physics

Title: Power Law in a Bounded Range: Estimating the Lower and Upper Bounds from Sample Data

Abstract: Power law distributions are widely observed in chemical physics, geophysics, biology, and beyond. The independent variable x of these distributions has an obligatory lower bound and in many cases also an upper bound. Estimating these bounds from sample data is notoriously difficult, with a recent method involving O(N^3) operations, where N denotes sample size. Here I develop an approach for estimating the lower and upper bounds that involves O(N) operations. The approach centers on calculating the mean values, x_min and x_max, of the smallest x and the largest x in N-point samples. A fit of x_min or x_max as a function of N yields the estimate for the lower or upper bound. Application to synthetic data demonstrates the accuracy and reliability of this approach.
Comments: 10 pages, 3 figures
Subjects: Chemical Physics (physics.chem-ph); Biological Physics (physics.bio-ph); Data Analysis, Statistics and Probability (physics.data-an); Geophysics (physics.geo-ph); Biomolecules (q-bio.BM)
Cite as: arXiv:2303.13456 [physics.chem-ph]
  (or arXiv:2303.13456v1 [physics.chem-ph] for this version)

Submission history

From: Huan-Xiang Zhou [view email]
[v1] Thu, 23 Mar 2023 17:24:54 GMT (1325kb)

Link back to: arXiv, form interface, contact.