ESS guidelines on temporal disaggregation by Dario Buono dario ESS guidelines on temporal disaggregation by Dario Buono dario.buono@ec.europa.eu (Eurostat, Unit B1 Methodology) National Accounts Working Group Luxembourg, 10-11 May 2017 Item 19
ESS Guidelines on TD Why? What for? How? Methodology harmonization: the definition of best practices What for? Deriving consistent high (Q/M) frequency data from low frequency (A/Q) data How? Estimation methods with quality measures
ESS Guidelines on TD Who? When? Where? All infra-annual data producers Regular production and revision analysis Where? Stand-alone and remote solution
Framework The ESS Guidelines on TD present both theoretical aspects and practical implementation issues in a user-friendly way Methods Specific domains Revisions Quality and documentation. CoP requirements are satisfied principle 7 (Sound Methodology) principles 14 (Coherence and Comparability) 15 (Accessibility and Clarity)
Approach 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 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 4 – SPECIFIC DOMAINS 5 - REVISION POLICIES 6 - ACCURACY OF BENCKMARKING AND RECONCILIATION
Item 4.1: Benchmarking and reconciliation in National Account QNA variables must be temporally consistent with annual benchmarks and respect the contemporaneous constraint (aggregates vs. components). This is very much demanded by users. Statistical methods for deriving QNA estimates are: Methods when only annual estimates are available, Methods that uses proxy indicators Available methods: Chow and Lin and variants Denton and variants
Roadmap 2017 March, SA Expert Group, some NSIs & NCBs 2017 April, WG Methodology Discussion (all MS) 2017 May, WG National Accounts (all MS) 2017 May-September, dedicated Task Force 2018 February, DIME/ITDG 2018 ESSC endorsement
Current Member of the TF on Temporal Disaggregation Eurostat: Dario Buono (Methodology), Riccardo Gatto (Labour Market), Enrico Infante (National Accounts) AT: Markus Fröhlich (Methodology) CZ: Karel Safr (National Accounts), Ondrej Vozar (Methodology) DK: Peter Stoltze (Methodology) IT: Barbara Gualdabascio (Methodology) NL: Mark Van Der Loo (Methodology) SE: Suad Elezovic (Methodology) UK: Duncan Elliot (Methodology)
…open points to kick-off the discussion Any suggestion for the overall structure? Is it needed to have devoted chapter for National Accounts and Labour Market? Should we include higher frequency data (e.g. weekly or daily)? What about the chapters on metadata and data presentation issues? Any comment about the part on revisions? Shall we quote the tools to be used? Interested in joining the Task Force?