Abstract:A primary motivation of this contribution is to define new locally stationary Markov models for categorical or integer-valued data. For this initial purpose, we propose a new general approach for dealing with time-inhomogeneity that extends the local stationarity notion developed in the time series literature. We also introduce a probabilistic framework which is very flexible and allows us to consider a much larger class of Markov chain models on arbitrary state spaces, including most of the locally stationary autoregressive processes studied in the literature. We consider triangular arrays of time-inhomogeneous Markov chains, defined by some families of contracting and slowly-varying Markov kernels. The finite-dimensional distribution of such Markov chains can be approximated locally with the distribution of ergodic Markov chains and some mixing properties are also available for these triangular arrays. As a consequence of our results, some classical geometrically ergodic homogeneous Markov chain models have a locally stationary version, which lays the theoretical foundations for new statistical modeling. Statistical inference of finite-state Markov chains can be based on kernel smoothing and we provide a complete and fast implementation of such models, directly usable by the practitioners. We also illustrate the theory on a real data set. A central limit theorem for Markov chains on more general state spaces is also provided and illustrated with the statistical inference in INAR models, Poisson ARCH models and binary time series models. Additional examples such as locally stationary regime-switching or SETAR models are also discussed.

Un Modèle FARIMA Localement Stationnaire

A local stationary long-memory model for internet traffic.

A Selection Criterion for Piecewise Stationary Long-Memory Models

Modeling Non-Stationary Long-Memory Signals with Large Amounts of Data

Piecewise Farima Models for Long-Memory Time Series

Structural Breaks Estimation for Long Memory Signals

Structural Changes Estimation for Strongly Dependent Processes

An empirical mode decomposition-based frequency-domain approach for the fatigue analysis of nonstationary processes

A Procedure for Modeling Non-Stationary Signals with Long Range Dependence

Modelling piecewise long memory signals based on MDL

Modeling non-stationarities in high-frequency financial time series

A Time-Varying FARIMA Model for Internet Traffic

A model-free characterization of recurrences in stationary time series

Modeling for a Class of Nonstationary ARMA Series 1

A Bayesian Method of Change-Point Estimation with Recurrent Regimes: Application to GARCH Models

Seasonal fractional long-memory processes. A semiparametric estimation approach

Dependent time changed processes with applications to nonlinear ocean waves

On-Line Structural Breaks Estimation for Non-stationary Time Series Models

Local stationarity and time-inhomogeneous Markov chains

The Nonparametric Estimation of Long Memory Spatio-Temporal Random Field Models