English

Tomography of scaling

Physics and Society 2019-10-16 v2 Disordered Systems and Neural Networks Data Analysis, Statistics and Probability

Abstract

Scaling describes how a given quantity YY that characterizes a system varies with its size PP. For most complex systems it is of the form YPβY\sim P^\beta with a nontrivial value of the exponent β\beta, usually determined by regression methods. The presence of noise can make it difficult to conclude about the existence of a non-linear behavior with β1\beta\neq 1 and we propose here to circumvent fitting problems by investigating how two different systems of sizes P1P_1 and P2P_2 are related to each other. This leads us to define a local scaling exponent βloc\beta_{\mathrm{loc}} that we study versus the ratio P2/P1P_2/P_1 and provides some sort of `tomography scan' of scaling across different values of the size ratio, allowing us to assess the relevance of nonlinearity in the system and to identify an effective exponent that minimizes the error for predicting the value of YY. We illustrate this method on various real-world datasets for cities and show that our method reinforces in some cases the standard analysis, but is also able to provide new insights in inconclusive cases and to detect problems in the scaling form such as the absence of a single scaling exponent or the presence of threshold effects.

Keywords

Cite

@article{arxiv.1908.11549,
  title  = {Tomography of scaling},
  author = {Marc Barthelemy},
  journal= {arXiv preprint arXiv:1908.11549},
  year   = {2019}
}

Comments

Revised version with additional results and discussions. 16 pages, 2 tables, 19 figures

R2 v1 2026-06-23T11:00:38.431Z