Statistical Signatures of Structural Organization: The case of long memory in renewal processes
arXiv:1512.01859 · doi:10.1016/j.physleta.2016.02.052
Abstract
Identifying and quantifying memory are often critical steps in developing a mechanistic understanding of stochastic processes. These are particularly challenging and necessary when exploring processes that exhibit long-range correlations. The most common signatures employed rely on second-order temporal statistics and lead, for example, to identifying long memory in processes with power-law autocorrelation function and Hurst exponent greater than . However, most stochastic processes hide their memory in higher-order temporal correlations. Information measures---specifically, divergences in the mutual information between a process' past and future (excess entropy) and minimal predictive memory stored in a process' causal states (statistical complexity)---provide a different way to identify long memory in processes with higher-order temporal correlations. However, there are no ergodic stationary processes with infinite excess entropy for which information measures have been compared to autocorrelation functions and Hurst exponents. Here, we show that fractal renewal processes---those with interevent distribution tails ---exhibit long memory via a phase transition at . Excess entropy diverges only there and statistical complexity diverges there and for all . When these processes do have power-law autocorrelation function and Hurst exponent greater than , they do not have divergent excess entropy. This analysis breaks the intuitive association between these different quantifications of memory. We hope that the methods used here, based on causal states, provide some guide as to how to construct and analyze other long memory processes.
arXiv comment13 pages, 2 figures, 3 appendixes; http://csc.ucdavis.edu/~cmg/compmech/pubs/lrmrp.htm
References in corpus (16)
- Are biological systems poised at criticality?
- Computational Mechanics: Pattern and Prediction, Structure and Simplicity
- Monotonous continuous-time random walks with drift and stochastic reset events
- Time's Barbed Arrow: Irreversibility, Crypticity, and Stored Information
- Prediction, Retrodiction, and The Amount of Information Stored in the Present
- Brain, Music and non-Poisson Renewal Processes
- Inferring hidden Markov models from noisy time sequences: a method to alleviate degeneracy in molecular dynamics
- Many Roads to Synchrony: Natural Time Scales and Their Algorithms
- Informational and Causal Architecture of Discrete-Time Renewal Processes
- Collective Phenomena and Non-Finite State Computation in a Human Social System
- Information Anatomy of Stochastic Equilibria
- Characterization of intermittency in renewal processes: Application to earthquakes
- On Hidden Markov Processes with Infinite Excess Entropy
- Group Minds and the Case of Wikipedia
- Enumerating Finitary Processes
- Bursting transition in a linear self-exciting point process
Cited by in corpus (10)
- Informational and Causal Architecture of Continuous-time Renewal and Hidden Semi-Markov Processes
- Nearly Maximally Predictive Features and Their Dimensions
- Characterising information-theoretic storage and transfer in continuous time processes
- Thermodynamics of Random Number Generation
- Ergodicity Breaking and Localization
- Conflict and Computation on Wikipedia: a Finite-State Machine Analysis of Editor Interactions
- Trimming the Independent Fat: Sufficient Statistics, Mutual Information, and Predictability from Effective Channel States
- Memory in quantum dot blinking
- The Origins of Computational Mechanics: A Brief Intellectual History and Several Clarifications
- Exploring Predictive States via Cantor Embeddings and Wasserstein Distance