Testing for news and noise in non-stationary time series subject to multiple historical revisions

Alain Hecq, Jan P. A. M. Jacobs*, Michalis P. Stamatogiannis

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

This paper focuses on testing non-stationary real-time data for forecastability, i.e., whether data revisions reduce noise or are news, by putting data releases in vector-error correction forms. To deal with historical revisions which affect the whole vintage of time series due to redefinitions, methodological innovations etc., we employ the recently developed impulse indicator saturation approach, which involves potentially adding an indicator dummy for each observation to the model. We illustrate our procedures with the U.S. real GNP/GDP series of the Federal Reserve Bank of Philadelphia and find that revisions to this series neither reduce noise nor can be considered as news.
Original languageEnglish
Pages (from-to)396-407
Number of pages12
JournalJournal of Macroeconomics
Volume60
DOIs
Publication statusPublished - Jun 2019

Keywords

  • Data revision
  • Cointegration
  • News-noise tests
  • Outlier detection
  • WEAK EXOGENEITY
  • VINTAGES
  • SELECTION
  • GROWTH
  • MODELS

Cite this

Hecq, Alain ; Jacobs, Jan P. A. M. ; Stamatogiannis, Michalis P. / Testing for news and noise in non-stationary time series subject to multiple historical revisions. In: Journal of Macroeconomics. 2019 ; Vol. 60. pp. 396-407.
@article{1debf5b30e164f1bacc4d35c5fd5d1ef,
title = "Testing for news and noise in non-stationary time series subject to multiple historical revisions",
abstract = "This paper focuses on testing non-stationary real-time data for forecastability, i.e., whether data revisions reduce noise or are news, by putting data releases in vector-error correction forms. To deal with historical revisions which affect the whole vintage of time series due to redefinitions, methodological innovations etc., we employ the recently developed impulse indicator saturation approach, which involves potentially adding an indicator dummy for each observation to the model. We illustrate our procedures with the U.S. real GNP/GDP series of the Federal Reserve Bank of Philadelphia and find that revisions to this series neither reduce noise nor can be considered as news.",
keywords = "Data revision, Cointegration, News-noise tests, Outlier detection, WEAK EXOGENEITY, VINTAGES, SELECTION, GROWTH, MODELS",
author = "Alain Hecq and Jacobs, {Jan P. A. M.} and Stamatogiannis, {Michalis P.}",
note = "data source:",
year = "2019",
month = "6",
doi = "10.1016/j.jmacro.2019.03.003",
language = "English",
volume = "60",
pages = "396--407",
journal = "Journal of Macroeconomics",
issn = "0164-0704",
publisher = "Elsevier Science",

}

Testing for news and noise in non-stationary time series subject to multiple historical revisions. / Hecq, Alain; Jacobs, Jan P. A. M.; Stamatogiannis, Michalis P.

In: Journal of Macroeconomics, Vol. 60, 06.2019, p. 396-407.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Testing for news and noise in non-stationary time series subject to multiple historical revisions

AU - Hecq, Alain

AU - Jacobs, Jan P. A. M.

AU - Stamatogiannis, Michalis P.

N1 - data source:

PY - 2019/6

Y1 - 2019/6

N2 - This paper focuses on testing non-stationary real-time data for forecastability, i.e., whether data revisions reduce noise or are news, by putting data releases in vector-error correction forms. To deal with historical revisions which affect the whole vintage of time series due to redefinitions, methodological innovations etc., we employ the recently developed impulse indicator saturation approach, which involves potentially adding an indicator dummy for each observation to the model. We illustrate our procedures with the U.S. real GNP/GDP series of the Federal Reserve Bank of Philadelphia and find that revisions to this series neither reduce noise nor can be considered as news.

AB - This paper focuses on testing non-stationary real-time data for forecastability, i.e., whether data revisions reduce noise or are news, by putting data releases in vector-error correction forms. To deal with historical revisions which affect the whole vintage of time series due to redefinitions, methodological innovations etc., we employ the recently developed impulse indicator saturation approach, which involves potentially adding an indicator dummy for each observation to the model. We illustrate our procedures with the U.S. real GNP/GDP series of the Federal Reserve Bank of Philadelphia and find that revisions to this series neither reduce noise nor can be considered as news.

KW - Data revision

KW - Cointegration

KW - News-noise tests

KW - Outlier detection

KW - WEAK EXOGENEITY

KW - VINTAGES

KW - SELECTION

KW - GROWTH

KW - MODELS

U2 - 10.1016/j.jmacro.2019.03.003

DO - 10.1016/j.jmacro.2019.03.003

M3 - Article

VL - 60

SP - 396

EP - 407

JO - Journal of Macroeconomics

JF - Journal of Macroeconomics

SN - 0164-0704

ER -