ITEM 3.2 ESS GUIDELINES ON TEMPORAL DISAGGREGATION BY DARIO BUONO (EUROSTAT) WG METHODOLOGY 5 APRIL 2017 FOREWORDS The Guidelines cover all issues related to Temporal
Disaggregation (TD) from annual to quarterly to monthly data, from choice of the methods, through the practice in specific domain, to revisions, quality measurement and documentation. The definition of best practices for TD postponed for a long time since the needs of the methodology harmonization for
the indicators which TD mainly apply. The ESS Guidelines on TD present both theoretical aspects and practical implementation issues in a friendly and easy to read framework. Requirements of principle 7 (Sound Methodology) of the ESS Code of Practice, principles 14 (Coherence and Comparability) and 15 (Accessibility and Clarity) are satisfied. Scope of guidelines
Common Guidelines for Temporal Disaggregation within the ESS as essential step for a better harmonization and comparability of infra-annual statistics Each stage of Benchmarking and Reconciliation (B&R) process is explained and options described. Out of these options three alternative courses of action are highlighted:
(A) Best alternative (B) Acceptable (C) To be avoided Guidelines summary PRINCIPLES FOR BENCKMARKING AND RECONCILIATION 1 - POLICY FOR BENCKMARKING AND RECONCILIATION 2 - TEMPORAL DISAGGREGATION
3 - SPECIFIC ISSUES ON BENCKMARKING AND RECONCILIATION Guidelines summary 4 SPECIFIC DOMAINS 5 - REVISION POLICIES 6 - ACCURACY OF BENCKMARKING AND RECONCILIATION
A POLICY FOR BENCKMARKING AND RECONCILIATION Item 1.1: Principles for benchmarking and reconciliation Item 1.2: The need for domain specific benchmarking and reconciliation policies Item 1.3: Consistency across domain specific benchmarking and reconciliation policies
Item 1.4: Stability of benchmarking and reconciliation policies TEMPORAL DISAGGREGATION Item 2.1 Temporal disaggregation Item 2.2: Alternative methods for benchmarking Item 2.3: Alternative methods for reconciliation Item 2.4: Choice of the software SPECIFIC ISSUES ON BENCKMARKING AND
RECONCILIATION Item 3.1: Mathematical vs Statistical methods Item 3.2: End of series Item 3.3: Seasonal Adjustment Item 3.4: Flash estimates Item 3.5: Outliers identification and treatment Item 3.6: Short time series Item 3.7: Chain linked series Item 3.8: Selection on one or more proxies
SPECIFIC DOMAIN Item 4.1: Quarterly National Account Item 4.2: Labour Market REVISION POLICIES Item 5.1: General revision policy and release calendar ACCURACY OF BENCKMARKING AND RECONCILIATION
Item 6.1: Accuracy policy for benchmarking and reconciliation Open points High frequency data (weekly or daily) A detailed procedure description for some relevant domains such as QNA o Labor Market Documenting metadata
Data presentation issues Cite the existing tools/plug-in Roadmap 2017 March, SAEG Initial discussion (some NSIs and NCBs) 2017 April, WG Methodology Discussion (all MS) 2017 May, WG National Accounts (all MS) 2017q3, revised version 2017q4, written consultation
2018, February, DIME/ITDG 2018, ESSC endorsement Main references Dagum, E.B., and P.A. Cholette (2006), Benchmarking, Temporal Disaggregation, and Reconciliation Methods for Time Series, Springer edition. IMF (2014) Quarterly National Accounts Manual - Concepts, Data Sources, and Compilation, chp.6 Benchmarking and Reconciliation
Di Fonzo, T., and M. Marini (2012a), Benchmarking Time Series according to a Growth Rates Preservation Principle, Journal of Economic and Social Measurement, Vol. 37(3), pp. 225252. Main references Di Fonzo, T., and Marini M. (2012b), On the Extrapolation with the Denton Proportional Benchmarking Method, IMF Working Paper Series, WP/12/169. ESS Guidelines on Seasonal Adjustment (2015) ESS Guidelines on Revision Policy (2013)
ESS Handbook on Quarterly National Accounts (2013)
