Why FARIMA models are brittle

DOI10.1142/S0218348X13500126MaRDI QIDQ2865149zbMATH OpenOpenAlexFDO

Authors D. Veitch, A. Gorst-Rasmussen, A. Gefferth

Publication date 28 November 2013

Published in Fractals (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1203.6140

zbMATH Keywords

long-range dependenceHurst parameterself-similarityfractional Gaussian noiseFARIMAfractionally differenced process

Mathematics Subject Classification ID

Fractional processes, including fractional Brownian motion (60G22) General second-order stochastic processes (60G12) Self-similar stochastic processes (60G18)

Abstract: The FARIMA models, which have long-range-dependence (LRD), are widely used in many areas. Through deriving a precise characterisation of the spectrum, autocovariance function, and variance time function, we show that this family is very atypical among LRD processes, being extremely close to the fractional Gaussian noise in a precise sense. Furthermore, we show that this closeness property is not robust to additive noise. We argue that the use of FARIMA, and more generally fractionally differenced time series, should be reassessed in some contexts, in particular when convergence rate under rescaling is important and noise is expected.

Recommendations

• Estimating FARIMA models with uncorrelated but non-independent error terms
• Piecewise FARIMA models for long-memory time series
• scientific article; zbMATH DE number 1301878
• scientific article; zbMATH DE number 727022
• ON GENERALIZED FRACTIONAL PROCESSES

Cites work

• AN INTRODUCTION TO LONG-MEMORY TIME SERIES MODELS AND FRACTIONAL DIFFERENCING
• An exponential model for the spectrum of a scalar time series
• An extension of a logarithmic form of Cramér's ruin theorem to some FARIMA and related processes
• ESTIMATORS FOR LONG-RANGE DEPENDENCE: AN EMPIRICAL STUDY
• Fitting long-memory models by generalized linear regression
• Fractional differencing
• Long-range Dependence: Revisiting Aggregation with Wavelets
• RATE OPTIMAL SEMIPARAMETRIC ESTIMATION OF THE MEMORY PARAMETER OF THE GAUSSIAN TIME SERIES WITH LONG‐RANGE DEPENDENCE
• Semiparametric analysis of long-memory time series
• The nature of discrete second-order self-similarity

Cited in

(4)

• Detecting long-range dependence for time-varying linear models
• FARIMA with stable innovations model of Great Salt Lake elevation time series
• An approximate fractional Gaussian noise model with \(\mathcal{O}(n)\) computational cost
• Estimating FARIMA models with uncorrelated but non-independent error terms

This page was built for publication: Why FARIMA models are brittle

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2865149)