Generalized Taylor's law for infinite-mean heavy-tailed data under dependence and heterogeneity.

Generalized Taylor's law applies to dependent, heterogeneous heavy-tailed data with infinite mean and variance, extending its use to complex systems.

• Why it matters: Understanding fluctuation scaling in such data is crucial because many real-world systems exhibit dependence and heterogeneity, challenging traditional assumptions of independence and finite moments.
• What they did: The authors developed a probabilistic framework for Taylor's law in heavy-tailed distributions, analyzing convergence rates and relaxing the independence assumption through theoretical and simulation studies.
• The result: This work broadens the applicability of Taylor's law to dependent time series and network data, enabling more accurate modeling and analysis of complex, heavy-tailed phenomena.