A Dynamic Taylor's law. (June 2022)

• Record Type: Journal Article
• Authors: De La Pena, Victor; Doukhan, Paul; Salhi, Yahia
• Is Part Of: Journal of applied probability. Volume 59:Number 2(2022)
• Page Start: 584
• Page End: 607
• Publication Date: 2022-06
• Subjects: Self-normalized sums -- Taylor's law -- weak dependence -- central limit theorem; 60G99 -- 60F05 -- 62P12; 519.2
• DOI: 10.1017/jpr.2021.40
• Languages: English
• ISSNs: 0021-9002

Abstract

Abstract: Taylor's power law (or fluctuation scaling) states that on comparable populations, the variance of each sample is approximately proportional to a power of the mean of the population. The law has been shown to hold by empirical observations in a broad class of disciplines including demography, biology, economics, physics, and mathematics. In particular, it has been observed in problems involving population dynamics, market trading, thermodynamics, and number theory. In applications, many authors consider panel data in order to obtain laws of large numbers. Essentially, we aim to consider ergodic behaviors without independence. We restrict our study to stationary time series, and develop different Taylor exponents in this setting. From a theoretical point of view, there has been a growing interest in the study of the behavior of such a phenomenon. Most of these works focused on the so-called static Taylor's law related to independent samples. In this paper we introduce a dynamic Taylor's law for dependent samples using self-normalized expressions involving Bernstein blocks. A central limit theorem (CLT) is proved under either weak dependence or strong mixing assumptions for the marginal process. The limit behavior of the estimation involves a series of covariances, unlike the classic framework where the limit behavior involves the marginal variance. We also provide an asymptotic result for a goodness-of-fit procedure suitable for checking whether the corresponding dynamic Taylor's law holds in empirical studies.