The effect of round-off error on long memory processes

Fabrizio Lillo, Gabriele La Spada, Fabrizio Lillo

Risultato della ricerca: Article

1 Citazione (Scopus)

Abstract

We study how the round-off (or discretization) error changes the statistical properties of a Gaussian long memory process. We show that the autocovariance and the spectral density of the discretized process are asymptotically rescaled by a factor smaller than one, and we compute exactly this scaling factor. Consequently, we find that the discretized process is also long memory with the same Hurst exponent as the original process. We consider the properties of two estimators of the Hurst exponent, namely the local Whittle (LW) estimator and the detrended fluctuation analysis (DFA). By using analytical considerations and numerical simulations we show that, in presence of round-off error, both estimators are severely negatively biased in finite samples. Under regularity conditions we prove that the LW estimator applied to discretized processes is consistent and asymptotically normal. Moreover, we compute the asymptotic properties of the DFA for a generic (i.e., non-Gaussian) long memory process and we apply the result to discretized processes.
Lingua originaleEnglish
pagine (da-a)445-482
Numero di pagine38
RivistaStudies in Nonlinear Dynamics and Econometrics
Volume18
Stato di pubblicazionePublished - 2014

Fingerprint

Long Memory Process
Rounding error
Hurst Exponent
Estimator
Fluctuations
fluctuation
Autocovariance
Scaling Factor
Discretization Error
Long Memory
Error Estimator
Spectral Density
Regularity Conditions
Gaussian Process
Statistical property
Asymptotic Properties
Biased
Long memory process
regularity
scaling

All Science Journal Classification (ASJC) codes

  • Analysis
  • Economics and Econometrics
  • Social Sciences (miscellaneous)

Cita questo

The effect of round-off error on long memory processes. / Lillo, Fabrizio; La Spada, Gabriele; Lillo, Fabrizio.

In: Studies in Nonlinear Dynamics and Econometrics, Vol. 18, 2014, pag. 445-482.

Risultato della ricerca: Article

Lillo, Fabrizio ; La Spada, Gabriele ; Lillo, Fabrizio. / The effect of round-off error on long memory processes. In: Studies in Nonlinear Dynamics and Econometrics. 2014 ; Vol. 18. pagg. 445-482.
@article{d045649b096b4df8bfa88e2a2127c238,
title = "The effect of round-off error on long memory processes",
abstract = "We study how the round-off (or discretization) error changes the statistical properties of a Gaussian long memory process. We show that the autocovariance and the spectral density of the discretized process are asymptotically rescaled by a factor smaller than one, and we compute exactly this scaling factor. Consequently, we find that the discretized process is also long memory with the same Hurst exponent as the original process. We consider the properties of two estimators of the Hurst exponent, namely the local Whittle (LW) estimator and the detrended fluctuation analysis (DFA). By using analytical considerations and numerical simulations we show that, in presence of round-off error, both estimators are severely negatively biased in finite samples. Under regularity conditions we prove that the LW estimator applied to discretized processes is consistent and asymptotically normal. Moreover, we compute the asymptotic properties of the DFA for a generic (i.e., non-Gaussian) long memory process and we apply the result to discretized processes.",
author = "Fabrizio Lillo and {La Spada}, Gabriele and Fabrizio Lillo",
year = "2014",
language = "English",
volume = "18",
pages = "445--482",
journal = "Studies in Nonlinear Dynamics and Econometrics",
issn = "1081-1826",
publisher = "Berkeley Electronic Press",

}

TY - JOUR

T1 - The effect of round-off error on long memory processes

AU - Lillo, Fabrizio

AU - La Spada, Gabriele

AU - Lillo, Fabrizio

PY - 2014

Y1 - 2014

N2 - We study how the round-off (or discretization) error changes the statistical properties of a Gaussian long memory process. We show that the autocovariance and the spectral density of the discretized process are asymptotically rescaled by a factor smaller than one, and we compute exactly this scaling factor. Consequently, we find that the discretized process is also long memory with the same Hurst exponent as the original process. We consider the properties of two estimators of the Hurst exponent, namely the local Whittle (LW) estimator and the detrended fluctuation analysis (DFA). By using analytical considerations and numerical simulations we show that, in presence of round-off error, both estimators are severely negatively biased in finite samples. Under regularity conditions we prove that the LW estimator applied to discretized processes is consistent and asymptotically normal. Moreover, we compute the asymptotic properties of the DFA for a generic (i.e., non-Gaussian) long memory process and we apply the result to discretized processes.

AB - We study how the round-off (or discretization) error changes the statistical properties of a Gaussian long memory process. We show that the autocovariance and the spectral density of the discretized process are asymptotically rescaled by a factor smaller than one, and we compute exactly this scaling factor. Consequently, we find that the discretized process is also long memory with the same Hurst exponent as the original process. We consider the properties of two estimators of the Hurst exponent, namely the local Whittle (LW) estimator and the detrended fluctuation analysis (DFA). By using analytical considerations and numerical simulations we show that, in presence of round-off error, both estimators are severely negatively biased in finite samples. Under regularity conditions we prove that the LW estimator applied to discretized processes is consistent and asymptotically normal. Moreover, we compute the asymptotic properties of the DFA for a generic (i.e., non-Gaussian) long memory process and we apply the result to discretized processes.

UR - http://hdl.handle.net/10447/107459

UR - http://www.degruyter.com/view/j/snde

M3 - Article

VL - 18

SP - 445

EP - 482

JO - Studies in Nonlinear Dynamics and Econometrics

JF - Studies in Nonlinear Dynamics and Econometrics

SN - 1081-1826

ER -