Testing for News and Noise in Non-Stationary Time Series Subject to Multiple Historical Revisions

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.

[1]  Jurgen A. Doornik,et al.  Empirical Model Discovery and Theory Evaluation: Automatic Selection Methods in Econometrics , 2014 .

[2]  Kerry Patterson,et al.  Exploiting information in vintages of time-series data , 2003 .

[3]  Rational Economic Data Revisions , 1987 .

[4]  Alain Hecq,et al.  Combining forecasts from successive data vintages: An application to U.S. growth , 2016 .

[5]  Dean Croushore,et al.  Forecasting with Real-Time Data Vintages , 2011 .

[6]  K. Patterson.  Which vintage of data to use when there are multiple vintages of data?: Cointegration, weak exogeneity and common factors , 2000 .

[7]  Neil R. Ericsson,et al.  Constructive Data Mining: Modeling Argentine Broad Money Demand , 2008 .

[8]  Andrew J. Patton,et al.  Predictability of Output Growth and Inflation: A Multi-Horizon Survey Approach , 2011 .

[9]  Michael P. Clements,et al.  Real-time forecasting of inflation and output growth with autoregressive models in the presence of data revisions , 2013 .

[10]  David I. Harvey.  The evaluation of economic forecasts , 1997 .

[11]  David E. Runkle,et al.  Are preliminary announcements of the money stock rational forecasts , 1984 .

[12]  D. Hendry,et al.  Regression Models with Data-Based Indicator Variables , 2005 .

[13]  N. Gregory Mankiw,et al.  News or Noise : An Analysis of GNP Revisions , 2007 .

[14]  Jurgen A. Doornik,et al.  Model selection when there are multiple breaks , 2012 .

[15]  Alain Hecq.  On the VAR-VECM Representation of Real Time Data , 2009 .

[16]  David F. Hendry,et al.  Empirical Model Discovery and Theory Evaluation , 2010 .

[17]  Simon van Norden,et al.  Why are initial estimates of productivity growth so unreliable , 2016 .

[18]  J. Mincer,et al.  Economic Forecasts and Expectations: Analysis of Forecasting Behavior and Performance , 1970 .

[19]  Dean D. Croushore,et al.  Frontiers of Real-Time Data Analysis , 2008 .

[20]  Jesus Gonzalo,et al.  Estimation of Common Long-Memory Components in Cointegrated Systems , 1995 .

[21]  Athanasios Orphanides,et al.  The Unreliability of Output-Gap Estimates in Real Time , 2002, Review of Economics and Statistics.

[22]  Dennis J. Fixler,et al.  News, Noise, and Estimates of the "True" Unobserved State of the Economy , 2013 .

[23]  Michael P. Clements,et al.  The Oxford handbook of economic forecasting , 2011 .

[24]  Neil R. Ericsson,et al.  Evaluating a Global Vector Autoregression for Forecasting , 2012, International Advances in Economic Research.

[25]  Hans-Eggert Reimers,et al.  Dealing with Benchmark Revisions in Real-Time Data: The Case of German Production and Orders Statistics , 2009 .

[26]  Jan P. A. M. Jacobs,et al.  Modeling data revisions: Measurement error and dynamics of 'true' values , 2011 .

[27]  George Kapetanios,et al.  A State Space Approach to Extracting the Signal From Uncertain Data , 2007 .

[28]  Jonas Schmitt,et al.  Likelihood Based Inference In Cointegrated Vector Autoregressive Models , 2016 .

[29]  David F. Hendry,et al.  Automatic selection of indicators in a fully saturated regression , 2008, Comput. Stat..

[30]  Jonathan H. Wright,et al.  Federal Reserve Board , 2000 .

[31]  Norman R. Swanson,et al.  Are Statistical Reporting Agencies Getting It Right? Data Rationality and Business Cycle Asymmetry , 2001 .

[32]  J. Urbain.  Weak Exogeneity in Error Correction Models , 1993 .

[33]  Evan F. Koenig,et al.  VAR Estimation and Forecasting When Data Are Subject to Revision , 2010 .

[34]  S. B. Aruoba,et al.  Data Revisions are Not Well-Behaved , 2004 .

[35]  Anthony Garratt,et al.  Real-Time Representations of the Output Gap , 2008, The Review of Economics and Statistics.

[36]  Anthony Garratt,et al.  Real time representation of the UK output gap in the presence of model uncertainty , 2009 .

[37]  J. Urbain.  Partial versus full system modelling of cointegrated systems an empirical illustration , 1995 .

[38]  K. D. Patterson.  Modelling the data measurement process for the index of production , 2002 .