Not all estimators are born equal: The empirical properties of some estimators of long memory

We present the results of a simulation study into the properties of 12 different estimators of the Hurst parameter, H, or the fractional integration parameter, d, in long memory time series which are available in R packages. We compare and contrast their performance on simulated Fractional Gaussian...

Full description

Bibliographic Details
Main Authors: Rea, W., Oxley, Leslie, Reale, M., Brown, J.
Format: Journal Article
Published: 2012
Online Access:http://hdl.handle.net/20.500.11937/25456
_version_ 1848751714317893632
author Rea, W.
Oxley, Leslie
Reale, M.
Brown, J.
author_facet Rea, W.
Oxley, Leslie
Reale, M.
Brown, J.
author_sort Rea, W.
building Curtin Institutional Repository
collection Online Access
description We present the results of a simulation study into the properties of 12 different estimators of the Hurst parameter, H, or the fractional integration parameter, d, in long memory time series which are available in R packages. We compare and contrast their performance on simulated Fractional Gaussian Noises and fractionally integrated series with lengths between 100 and 10,000 data points and H values between 0.55 and 0.90 or d values between 0.05 and 0.40. We apply all 12 estimators to the Campito Mountain data and estimate the accuracy of their estimates using the Beran goodness-of-fit test for long memory time series. © 2012 IMACS.
first_indexed 2025-11-14T07:57:07Z
format Journal Article
id curtin-20.500.11937-25456
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T07:57:07Z
publishDate 2012
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-254562017-09-13T15:18:09Z Not all estimators are born equal: The empirical properties of some estimators of long memory Rea, W. Oxley, Leslie Reale, M. Brown, J. We present the results of a simulation study into the properties of 12 different estimators of the Hurst parameter, H, or the fractional integration parameter, d, in long memory time series which are available in R packages. We compare and contrast their performance on simulated Fractional Gaussian Noises and fractionally integrated series with lengths between 100 and 10,000 data points and H values between 0.55 and 0.90 or d values between 0.05 and 0.40. We apply all 12 estimators to the Campito Mountain data and estimate the accuracy of their estimates using the Beran goodness-of-fit test for long memory time series. © 2012 IMACS. 2012 Journal Article http://hdl.handle.net/20.500.11937/25456 10.1016/j.matcom.2012.08.005 restricted
spellingShingle Rea, W.
Oxley, Leslie
Reale, M.
Brown, J.
Not all estimators are born equal: The empirical properties of some estimators of long memory
title Not all estimators are born equal: The empirical properties of some estimators of long memory
title_full Not all estimators are born equal: The empirical properties of some estimators of long memory
title_fullStr Not all estimators are born equal: The empirical properties of some estimators of long memory
title_full_unstemmed Not all estimators are born equal: The empirical properties of some estimators of long memory
title_short Not all estimators are born equal: The empirical properties of some estimators of long memory
title_sort not all estimators are born equal: the empirical properties of some estimators of long memory
url http://hdl.handle.net/20.500.11937/25456