A note on Michelacci and Zaffaroni, long memory, and time series of economic growth

G. Silverberg, B. Verspagen

Research output: Book/ReportReportAcademic

23 Downloads (Pure)

Abstract

In a recent paper in The Journal of Monetary Economics, Michelacci and Zaffaroni (2000) estimate long memory parameters for GDP per capita of 16 OECD countries. In this note we argue that these estimations are questionable for the purposes of clarifying the time series properties of these data (presence of unit roots, mean reversion, long memory) because the authors a) filter out a deterministic linear-in-logs trend instead of first-differencing in logs, and manipulate the data in other highly questionable ways, b) rely on the semiparametric Geweke and Porter-Hudak (GPH) method as modified by Robinson, which is known to be highly biased in small samples. We re-examine these results using Beran’s nonparametric FGN estimator and Sowell’s exact maximum likelihood ARFIMA estimator. These methods avoid the small-sample bias and arbitrariness of the cut-off parameters of Robinson’s method and allow us to control for short memory effects, although the parametric ARFIMA estimator introduces specification problems of its own. We also look at the influence of the choice of sub-periods on the results. Finally, we apply Robinson’s method to our treatment of the data and show that MZ’s results no longer hold, nor are their cut-off parameter and filtering insensitivity claims substantiated.
Original languageEnglish
Place of PublicationEindhoven
PublisherEindhoven Center for Innovation Studies
Number of pages16
Publication statusPublished - 2000

Publication series

NameECIS working paper series
Volume200017

Fingerprint

economic growth
time series
OECD
Gross Domestic Product
filter
method
parameter

Cite this

Silverberg, G., & Verspagen, B. (2000). A note on Michelacci and Zaffaroni, long memory, and time series of economic growth. (ECIS working paper series; Vol. 200017). Eindhoven: Eindhoven Center for Innovation Studies.
Silverberg, G. ; Verspagen, B. / A note on Michelacci and Zaffaroni, long memory, and time series of economic growth. Eindhoven : Eindhoven Center for Innovation Studies, 2000. 16 p. (ECIS working paper series).
@book{9dfb778da1294bbfa79455386a3023ec,
title = "A note on Michelacci and Zaffaroni, long memory, and time series of economic growth",
abstract = "In a recent paper in The Journal of Monetary Economics, Michelacci and Zaffaroni (2000) estimate long memory parameters for GDP per capita of 16 OECD countries. In this note we argue that these estimations are questionable for the purposes of clarifying the time series properties of these data (presence of unit roots, mean reversion, long memory) because the authors a) filter out a deterministic linear-in-logs trend instead of first-differencing in logs, and manipulate the data in other highly questionable ways, b) rely on the semiparametric Geweke and Porter-Hudak (GPH) method as modified by Robinson, which is known to be highly biased in small samples. We re-examine these results using Beran’s nonparametric FGN estimator and Sowell’s exact maximum likelihood ARFIMA estimator. These methods avoid the small-sample bias and arbitrariness of the cut-off parameters of Robinson’s method and allow us to control for short memory effects, although the parametric ARFIMA estimator introduces specification problems of its own. We also look at the influence of the choice of sub-periods on the results. Finally, we apply Robinson’s method to our treatment of the data and show that MZ’s results no longer hold, nor are their cut-off parameter and filtering insensitivity claims substantiated.",
author = "G. Silverberg and B. Verspagen",
year = "2000",
language = "English",
series = "ECIS working paper series",
publisher = "Eindhoven Center for Innovation Studies",

}

Silverberg, G & Verspagen, B 2000, A note on Michelacci and Zaffaroni, long memory, and time series of economic growth. ECIS working paper series, vol. 200017, Eindhoven Center for Innovation Studies, Eindhoven.

A note on Michelacci and Zaffaroni, long memory, and time series of economic growth. / Silverberg, G.; Verspagen, B.

Eindhoven : Eindhoven Center for Innovation Studies, 2000. 16 p. (ECIS working paper series; Vol. 200017).

Research output: Book/ReportReportAcademic

TY - BOOK

T1 - A note on Michelacci and Zaffaroni, long memory, and time series of economic growth

AU - Silverberg, G.

AU - Verspagen, B.

PY - 2000

Y1 - 2000

N2 - In a recent paper in The Journal of Monetary Economics, Michelacci and Zaffaroni (2000) estimate long memory parameters for GDP per capita of 16 OECD countries. In this note we argue that these estimations are questionable for the purposes of clarifying the time series properties of these data (presence of unit roots, mean reversion, long memory) because the authors a) filter out a deterministic linear-in-logs trend instead of first-differencing in logs, and manipulate the data in other highly questionable ways, b) rely on the semiparametric Geweke and Porter-Hudak (GPH) method as modified by Robinson, which is known to be highly biased in small samples. We re-examine these results using Beran’s nonparametric FGN estimator and Sowell’s exact maximum likelihood ARFIMA estimator. These methods avoid the small-sample bias and arbitrariness of the cut-off parameters of Robinson’s method and allow us to control for short memory effects, although the parametric ARFIMA estimator introduces specification problems of its own. We also look at the influence of the choice of sub-periods on the results. Finally, we apply Robinson’s method to our treatment of the data and show that MZ’s results no longer hold, nor are their cut-off parameter and filtering insensitivity claims substantiated.

AB - In a recent paper in The Journal of Monetary Economics, Michelacci and Zaffaroni (2000) estimate long memory parameters for GDP per capita of 16 OECD countries. In this note we argue that these estimations are questionable for the purposes of clarifying the time series properties of these data (presence of unit roots, mean reversion, long memory) because the authors a) filter out a deterministic linear-in-logs trend instead of first-differencing in logs, and manipulate the data in other highly questionable ways, b) rely on the semiparametric Geweke and Porter-Hudak (GPH) method as modified by Robinson, which is known to be highly biased in small samples. We re-examine these results using Beran’s nonparametric FGN estimator and Sowell’s exact maximum likelihood ARFIMA estimator. These methods avoid the small-sample bias and arbitrariness of the cut-off parameters of Robinson’s method and allow us to control for short memory effects, although the parametric ARFIMA estimator introduces specification problems of its own. We also look at the influence of the choice of sub-periods on the results. Finally, we apply Robinson’s method to our treatment of the data and show that MZ’s results no longer hold, nor are their cut-off parameter and filtering insensitivity claims substantiated.

M3 - Report

T3 - ECIS working paper series

BT - A note on Michelacci and Zaffaroni, long memory, and time series of economic growth

PB - Eindhoven Center for Innovation Studies

CY - Eindhoven

ER -

Silverberg G, Verspagen B. A note on Michelacci and Zaffaroni, long memory, and time series of economic growth. Eindhoven: Eindhoven Center for Innovation Studies, 2000. 16 p. (ECIS working paper series).