Quality Reporting with JD+

Slides:



Advertisements
Similar presentations
Recent developments in the Business Tendency Survey area at the OECD Michela Gamba, Statistics Directorate OECD – EU Workshop on BTS – Brussels November.
Advertisements

- ONS Classification Coding Tools Project Occupation Classification Workshop RSS, London, 21 June 2004 Nigel Swier.
United Nations Economic Commission for Europe Statistical Division UNECE Training Workshop on Dissemination of MDG Indicators and Statistical Information.
Quality Guidelines for statistical processes using administrative data European Conference on Quality in Official Statistics Q2014 Giovanna Brancato, Francesco.
Regional Workshop for African Countries on Compilation of Basic Economic Statistics Pretoria, July 2007 Administrative Data and their Use in Economic.
Seasonal Adjustment Methods and Country Practices Based on the: Hungarian Central Statistical Office: Seasonal Adjustment Methods and Practices; UNECE.
J. Khélif Insee July 2008 A quality report for seasonally and trading day adjusted French IIP.
Demetra+ Quick Tour Versatile software. Choose the right tool Demetra+ main feature: multi-processing Demetra+ in production. Understanding.
REFERENCE METADATA FOR DATA TEMPLATE Ales Capek EUROSTAT.
USING DEMETRA+ IN DAILY WORK SAUG – Luxembourg, 16 October 2012 Enrico INFANTE, Eurostat Unit B1: Quality, Methodology and Research.
United Nations Economic Commission for Europe Statistical Division Seasonal Adjustment Process with Demetra+ Anu Peltola Economic Statistics Section, UNECE.
Ketty Attal-Toubert and Stéphanie Himpens Insee, France 16th of November, 2011 ESTP course Demetra+ Demetra+ for X12 in Daily Work.
for statistics based on multiple sources
1 Departamento de Contas Nacionais / Serviço de Indicadores de Curto Prazo National Accounts Department / Short Term Statistics Unit Using Demetra+
Ketty Attal-Toubert and Stéphanie Himpens Insee 22nd of June, 2011 An Overview of seasonal adjustment in the short term statistic department.
Outlining a Process Model for Editing With Quality Indicators Pauli Ollila (part 1) Outi Ahti-Miettinen (part 2) Statistics Finland.
Dominique Ladiray Gian Luigi Mazzi Q2008, Roma, 9-11 July 2008 Assessing the Quality of the Euro-Indicator Database.
April 2011 Testing Seasonal Adjustment with Demetra+ Ariunbold Shagdar National Statistical Office, Mongolia.
Presentation 1 / NBB report Towards JD+ 2.1…
Jean-Roch Vlimant, CERN Physics Performance and Dataset Project Physics Data & MC Validation Group McM : The Evolution of PREP. The CMS tool for Monte-Carlo.
Towards a seasonal adjustment and a revision policy Anu Peltola Economic Statistics Section, UNECE UNECE Workshop on Seasonal Adjustment 20 – 23 February.
General Recommendations on STS Carsten Boldsen Hansen Economic Statistics Section, UNECE UNECE Workshop on Short-Term Statistics (STS) and Seasonal Adjustment.
USING DEMETRA+ IN DAILY WORK SAUG – Luxembourg, 16 October 2012 Enrico INFANTE, Eurostat Unit B1: Quality, Methodology and Research.
Seasonal Adjustment Center of Excellence Quality Report for JD+ Towards a versatile quality report SAEG Meeting - December 7 th, 2015 SACE 1.
Ketty Attal-Toubert and Stéphanie Himpens Insee 22nd of June, 2011 Using SAS to implement additional tools.
13 November, 2014 Seminar on Quality Reports QUALITY REPORTS EXPERIENCE OF STATISTICS LITHUANIA Nadiežda Alejeva Head, Price Statistics.
Labour Cost Index (LCI) Calculation of the LCI in Denmark.
Labour Cost Index (LCI) European Union Regulations.
Quality declarations Study visit from Ukraine 19. March 2015
JDemetra+ as an innovative tool for seasonal adjustment
Implementation of Quality indicators for administrative data
Conference of European Statistics Stakeholders, 2016
FUTURE EVOLUTION OF SHORT-TERM ECONOMIC STATISTICS
Artur Andrysiak Economic Statistics Section, UNECE
Carsten Boldsen Hansen Economic Statistics Section, UNECE
Handbook on Seasonal Adjustment
2.1 Latest features of JDemetra+ 2.2
SAEG, 15 March 2018 Item 2.2 Development: progress and next steps. Contribution of the NBB
ESS guidelines on temporal disaggregation: Benchmarking and reconciliation From Annual to Quarterly to Monthly data.
MSDs and combined metadata reporting
Survey phases, survey errors and quality control system
SAEG 7th June 2016 Item 5.2 Eurostat migration to JD+: state of the art By Dario Buono.
(VIP-EDC) Point 6 of the agenda
Survey phases, survey errors and quality control system
SAEG 7th June 2016 Item 4.1 Eurostat work programme : priority areas-future Centre of Excellence – expected deliverables By Dario Buono.
2. An overview of SDMX (What is SDMX? Part I)
LAMAS Working Group January 2016
Hungarian practice on chain-linking and its implication for SA
Data Validation in the ESS Context
STATISTICAL AGENCY UNDER PRESIDENT OF THE REPUBLIC OF TAJIKISTAN
PRESENTATION OF SHORT-TERM ECONOMIC STATISTICS
Data Validation in the ESS Context
Sharing data validation activities in the ESS.
Item 3.2 ESS guidelines on temporal disaggregation by Dario Buono (Eurostat) WG Methodology 5 April 2017.
SAUG and SACE: Status Report
Mapping Data Production Processes to the GSBPM
Metadata used throughout statistics production
Faiz Alsuhail 21 of June, Frankfurt
Quality Reporting in CBS
Education and Training Statistics Working Group, May 2011
General Challenges and Drawbacks.
Template for methodological application
Measuring, reporting and communicating quality of National Accounts statistics (ESA 2010) in an integrated way with data production Christos LIOURIS,
Metadata on quality of statistical information
Prodcom Working Group Item Quality reporting and indicators
Latest features Frankfurt, 7 June 2016
Model Selection, Seasonal Adjustment, Analyzing Results
Testing seasonal adjustment with Demetra+
SILC draft implementing and delegated acts Item 3.4 of the agenda
ESS conceptual standards for quality reporting
Presentation transcript:

Quality Reporting with JD+ Where are we? Where are we going? SAEG – June 7th, 2016 SACE

The Current JD+ Quality Report: Principle n numerical quality indicators. Each of them is translated into a qualitative indicator. This set of qualitative indicators is then summarized in a qualitative global indicator. Use of colors: green, orange, red, severe Palate Jean (2014), JD+ Quality Indicators, NBB SACE

Current JD+ Quality Report SACE

JD+ QR: Selected Indicators Diagnostics on: The coherence of the decomposition ("Basic checks" group), Visual spectral inspection ("Visual spectral analysis"), The residuals of the RegArima pre-processing ("RegArima residuals" group), The residual seasonality ("Residual seasonality" group) The decomposition ("Seats" group for Tramo-Seats, M-Statistics group for X12). Most of them use parameters (usually thresholds) that can be modified Each group of diagnostics might be disabled, when it is considered as meaningless. SACE

Pros and Cons Pros Cons Very simple to read Very useful for the producer Cons Could be misunderstood (Warnings vs. Quality) Tramo/Seats “biased” Too many indicators (7 different seasonality tests) But output csv matrix incomplete SACE 5

The “Quality Report Plug-In Project” 2 very different ideas/objectives Technical side: Disseminating the knowledge on JD+ to make people able to contribute Statistical side: Developing a customized Quality Report The challenges Getting basic programming skills What is a Quality Report? Selecting and weighting statistical indicators SACE 6

Quality Report: A Quite Fuzzy Concept “Quality report” are words widely used in the SA context but they cover various meanings and concepts. When is it used, by whom, and for which purposes? Producing, Disseminating, Using (the series) SACE

Producing the series To define the process To run the process choosing the best strategy/model To run the process using the old/previous parameters To check/improve the process Annual campaign, production time To prepare the series for publication & storage SACE

Disseminating and storing the series Various publications Informing on overall quality and revisions etc. Various databases, various metadata Producer databases Institute databases Other databases Often a need for other quality diagnostics (timeliness for example) SACE

Using the series “Basic” user Researcher/Econometrician Direct use of the series (indexing contract) Use of the series to comment the evolution of the sector/Economy Usually “trust the series”; basic QR Researcher/Econometrician Focus on very specific criteria: timeliness, revisions, noise, dynamics of the series (model) etc. Use the QR ….. or not: redo its own adjustment SACE

Various objectives → Various QR? Yes Metadata ≠ Publication ≠ Producer ≠ Researcher …. and No Almost every information we can think about can be output by JD+. Metadata and Publication: “just a reporting problem” Some problems: JD+ cannot contain all necessary information (timeliness, real time revisions, classifications etc.) Quality of indirect aggregates (chain-linked or not) SACE

What can do JD+? Lots of things ….. But not everything The output matrix could be extended / customized The signal alerts (red lights) can be adapted/tuned Reports can be automatically created using standard formats (SDMX) But not everything It might consume resources (computing time, space) if too many indicators are required. JD+ does not store all the information useful for a QR JD+ does not handle “direct vs. indirect” aggregates. SACE

A Generic Structure A set of descriptive data (ex: method, model, parameters, length etc.) Weight of the descriptive data A set of quality statistics (ex: M1-M11 statistics in X-12) Weight of the statistics and quality statistics Weight of the series SACE

The “descriptive data” Typically information necessary for the metadata “Clarity” and “transparency” aspects of quality Information on the series (name, length etc.), on the process (method, regressors etc.) and on the result of the process (model, # of outliers, TD effect etc.) Could be also used to rerun the process Weight Typically, each item is present (1) or not present (0). Allows summarizing the overall quality (presence) of the “metadata” SACE

The “quality statistics” Typically information necessary to check the efficiency of the process for a series “accuracy” aspect of quality No residual seasonality, TD effect, M1-M11 statistics, JD+ set of quality statistics Also used to improve the seasonal adjustment Weight To derive a global quality index: Q-statistics, JD+ alerts. SACE

Weight of the series Important in production time Examples: weight of the sector in the NACE, contribution of the sector to the growth of the aggregate, weight of the country in the UE etc. Permits a selective editing: if something is going wrong during production, you focus on the most important series SACE

The real problems to solve The choice of the information to output Revisions: the last value of the series can be kept Some metadata cannot be included (see example) Too many indicators (7 seasonality tests) The choice of the weights Simple (?) for metadata (1/0) More complex for quality statistics How to use the QR in production time? SACE

A First Plug-In for Eurostat Unit F3 To report on the SA of Job Vacancy Survey series First specifications during a Hackathon session organized by the SAUG (08/12/2015) Several iterations to agree on the quality indicators A “final product” delivered end of 02/2016. Used in production by Member States SACE 18

The “command” SACE 19

The “final” version SACE 20

JD+ Quality Report Joint work with Istat and NBB First steps: Learning to develop simple plug-ins Output a very complete csv matrix for both TS & X12 First statistical studies using this basic plug-in to select relevant quality indicators Checking the probable redundancy of seasonality tests SACE 21