Processes with Long Memory: Regenerative Construction and Perfect Simulation

Abstract

We present a perfect simulation algorithm for stationary processes indexed by Z, with summable memory decay. Depending on the decay, we construct the process on finite or semi-infinite intervals, explicitly from an i.i.d. uniform sequence. Even though the process has infinite memory, its value at time 0 depends only on a finite, but random, number of these uniform variables. The algorithm is based on a recent regenerative construction of these measures by Ferrari, Maass, Mart{\'ı}nez and Ney. As applications, we discuss the perfect simulation of binary autoregressions and Markov chains on the unit interval.

27 pages, one figure. Version accepted by Annals of Applied Probability. Small changes with respect to version 2

Cited by in corpus (24)

• Infinite systems of interacting chains with memory of variable length - a stochastic model for biological neural nets
• Perfect simulation for interacting point processes, loss networks and Ising models
• Joint estimation of intersecting context tree models
• Rumor processes on $\mathbb{N}$ and discrete renewal processes
• Attractive regular stochastic chains: perfect simulation and phase transition
• Potts models with a defect line
• A stochastic system with infinite interacting components to model the time evolution of the membrane potentials of a population of neurons
• Developments in perfect simulation of Gibbs measures through a new result for the extinction of Galton-Watson-like processes
• On non-regular g-measures
• Kalikow-type decomposition for multicolor infinite range particle systems
• Stationary processes whose filtrations are standard
• Divergence rates of Markov order estimators and their application to statistical estimation of stationary ergodic processes
• Two-sided infinite-bin models and analyticity for Barak-Erdős graphs
• Chains of infinite order, chains with memory of variable length, and maps of the interval
• One-dimensional infinite memory imitation models with noise
• Estimation of the last passage percolation constant in a charged complete directed acyclic graph via perfect simulation
• Thermodynamic Formalism in Neuronal Dynamics and Spike Train Statistics
• Perfect simulation of autoregressive models with infinite memory
• Perfect simulation of a coupling achieving the $\bar{d}$-distance between ordered pairs of binary chains of infinite order
• Last passage percolation and limit theorems in Barak-Erdős directed random graphs and related models
• Stochastic processes with random contexts: a characterization, and adaptive estimators for the transition probabilities
• Sufficient conditions for the filtration of a stationary processes to be standard
• Freezing in the Infinite-Bin Model
• Ornstein-Zernike behavior for Ising models with infinite-range interactions