A maximum entropy framework for non-exponential distributions

Abstract

Probability distributions having power-law tails are observed in a broad range of social, economic, and biological systems. We describe here a potentially useful common framework. We derive distribution functions $\{p_k\}$ for situations in which a `joiner particle' $k$ pays some form of price to enter a `community' of size $k-1$, where costs are subject to economies-of-scale (EOS). Maximizing the Boltzmann-Gibbs-Shannon entropy subject to this energy-like constraint predicts a distribution having a power-law tail; it reduces to the Boltzmann distribution in the absence of EOS. We show that the predicted function gives excellent fits to 13 different distribution functions, ranging from friendship links in social networks, to protein-protein interactions, to the severity of terrorist attacks. This approach may give useful insights into when to expect power-law distributions in the natural and social sciences.

9 pages, 4 figures, 1 table

References in corpus (9)

• Power-law distributions in empirical data
• Kronecker Graphs: An Approach to Modeling Networks
• Preferential attachment in the growth of social networks: the case of Wikipedia
• Complex cooperative networks from evolutionary preferential attachment
• Power-Law Distributions for a Trapped Ion Interacting with a Classical Buffer Gas
• Toolbox model of evolution of prokaryotic metabolic networks and their regulation
• Nonuniversal power law scaling in the probability distribution of scientific citations
• Predictability of extreme events in social media
• Simulated evolution of protein-protein interaction networks with realistic topology

Cited by in corpus (4)

• Predictability of extreme events in social media
• A complete data frame work for fitting power law distributions
• Inferring transition rates on networks with incomplete knowledge
• Detecting temperature fluctuations at equilibrium