Using Demetra+ at NSO of FI, - first results of testing

Slides:



Advertisements
Similar presentations
Maine workforce conditions through September 2012 Presented to the Consensus Economic Forecasting Commission October 25, 2012 Glenn Mills Chief Economist.
Advertisements

Stockholm Economy Q Economic growth, private sector Gross pay, change (%) from the previous quarter, seasonally adjusted and trend-adjusted, annual.
Seasonal Adjustment of National Index Data at International Level
Using Demetra+ in daily work at NBP – SA in the time of crisis Sylwia Grudkowska, Department of Statistics, NBP.
Recessions and the Seasonal Adjustment of Industrial Production.
REPUBLIC OF TURKEY TURKISH STATISTICAL INSTITUTE TurkStat Direct vs. Indirect Approach in Seasonal Adjustment: Proposal for a new tool Necmettin Alpay.
Macroeconomic Facts Chapter 3. 2 Introduction Two kinds of regularities in economic data: -Relationships between the growth components in different variables.
X-12 ARIMA Eurostat, Luxembourg Seasonal Adjustment.
WORKING PARTY ON NATIONAL ACCOUNTS Paris, 3-5 October 2007 Revisions in Quarterly GDP of OECD Countries: An Update Document STD/CSTAT/WPNA(2007)15 Richard.
J. Khélif Insee July 2008 A quality report for seasonally and trading day adjusted French IIP.
Georgian Quarterly National Accounts and time-series of Short Term Statistics indicators Levan Gogoberishvili Head of National Accounts Division, Geostat.
United Nations Economic Commission for Europe Statistical Division Seasonal Adjustment Process with Demetra+ Anu Peltola Economic Statistics Section, UNECE.
Ronald van der Stegen Reducing statistical discrepancy between direct and indirect GDP.
Overview of Main Quality Diagnostics Anu Peltola Economic Statistics Section, UNECE UNECE Workshop on Seasonal Adjustment 20 – 23 February 2012, Ankara,
Ketty Attal-Toubert and Stéphanie Himpens Insee, France 16th of November, 2011 ESTP course Demetra+ Demetra+ for X12 in Daily Work.
1 Departamento de Contas Nacionais / Serviço de Indicadores de Curto Prazo National Accounts Department / Short Term Statistics Unit Using Demetra+
1 Forecasting with Intervention: Tourism in Croatia Ante Rozga, Toni Marasović, Josip Arnerić University of Split, Croatia.
Testing Seasonal Adjustment with Demetra+ Levan Gogoberishvili Head of National Accounts Division National Statistics Office of Georgia.
Ketty Attal-Toubert and Stéphanie Himpens Insee 22nd of June, 2011 An Overview of seasonal adjustment in the short term statistic department.
United Nations Economic Commission for Europe Statistical Division Importance of Original Data and Fixed Base Indices – Data Example Artur Andrysiak Economic.
Testing seasonal adjustment with Demetra+ Dovnar Olga Alexandrovna The National Statistical Committee, Republic of Belarus.
April 2011 Testing Seasonal Adjustment with Demetra+ Ariunbold Shagdar National Statistical Office, Mongolia.
Using Demetra+ at SURS Manca Golmajer, Andrejka Smukavec 20 June 2013.
Irwin/McGraw-Hill © Andrew F. Siegel, 1997 and l Chapter 14 l Time Series: Understanding Changes over Time.
SAEG 7 th Dec 2015 Item 2.5 Eurostat migration to JD+: state of the art By Dario Buono 1.
Testing Seasonal Adjustment of the Price Index for tomatoes with Demetra+ Kumpeisova Dinara Agency of Statistics of the Republic of Kazakhstan, Kazakhstan.
Ketty Attal-Toubert and Stéphanie Himpens Insee 22nd of June, 2011 Using SAS to implement additional tools.
——Comparing between the methods based on X-12ARIMA and TRAMO/SEATS By Ming-Fang Zhang Suo-Xian Pan IAOS Conference, Shanghai, October, 2008 Study.
September 2006OECD International Trade Statistics Experts Meeting OECD Main Economic Indicators Revisions Analysis Database Richard McKenzie OECD.
Update From WIOA Unified Plan-Economic Conditions
Inflation Report February 2006.
An Update on Business Employment Dynamics
Conference of European Statistics Stakeholders, 2016
Seasonal Adjustment Methods and Country Practices
Barbro Wickman-Parak, 3 October 2007
Carsten Boldsen Hansen Economic Statistics Section, UNECE
Shohreh Mirzaei Yeganeh United Nations Industrial Development
Change of turnover Source: Statistics Finland,
Revision Analysis for General Economic Statistics
TESTING AMI PERFORMANCE OF J-DEMETRA+ Turaç YAVUZ TURKSTAT
Henri Luomaranta, Statistics Finland
Testing seasonal adjustment with Demetra+
Macroeconomic heatmap taking the temperature of the Estonian economy
SHORT TERM INDICATORS IN THE SERVICE SECTOR
SA results obtained with Demetra2.2 and Demetra+
Difficulties in Seasonal Adjustment
How to select regressors and specifications in Demetra+?
Hungarian practice on chain-linking and its implication for SA
Seasonal adjustment with Demetra+
STATISTICAL AGENCY UNDER PRESIDENT OF THE REPUBLIC OF TAJIKISTAN
Testing seasonal adjustment with Demetra+
SA Economic Indicators
STATISTICS KAZAKHSTAN’S RESPONSE TO CHANGES IN ECONOMIC SITUATION MESHIMBAYEVA ANAR CHAIRPERSON, AGENCY OF THE REPUBLIC OF KAZAKHSTAN ON STATISTICS.
NATIONAL STATISTICAL INSTITUTE
Demetra+ User Manual new version
Faiz Alsuhail 21 of June, Frankfurt
LAMAS Working Group June 2015
Problematic time series and how to treat them
Testing seasonal adjustment using Demetra+
Task Force on Seasonal Adjustment of Quarterly National Accounts Second meeting Item 7 Calendar adjustment: Issues note Martin Eiglsperger Frankfurt,
Demetra+ User Manual Final release
Yankov V.M. Rosstat, Russian Federation
Faiz Alsuhail, Statistics Finland,
TYPES of seasonal adjustment processes in banco de españa
Short Term Statistics in National Accounts
Ermurachi Galina National Bureau of Statistics, Republic of Moldova
Summary of methods presented at the CMBF
Issues on Seasonal Adjustment in the EECCA countries
Using JDemetra+ at STATEC
Testing seasonal adjustment with Demetra+
Presentation transcript:

Using Demetra+ at NSO of FI, - first results of testing Faiz Alsuhail, faiz.alsuhail@stat.fi, +358 9 1734 2921

Introduction Statistics Finland currently uses Tramo/Seats and Demetra (v. 2.2) to run seasonal adjustment. This test is about using Tramo/Seats in J-Demetra+ (v. 1.2.0) . The question is whether the results (seasonally adjusted time series, trend, trading day corrected series) are the same in both software programs. Identical model settings are used. Faiz Alsuhail/ Statistics Finland

Time series studied We study five time series: Gross national product (GNP) Volume index for industrial output (VIIO), total industries Labor cost index (LCI) for all industries in private sector Unemployment rate for 15-74 year old Retail trade These time series are important, draw considerable attention from the users. They are seasonally adjusted in different ways. Faiz Alsuhail/ Statistics Finland

Time series studied Series Freq. Series span n° of obs. n° of TD regr. n° of outliers Transf. ARIMA-model GNP Quarterly 1990Q1 - 2012Q4 92 1 2 Log (0 1 1) (0 1 1) VIIO Monthly 1995M1 -2013M1 217 7 3 LCI 2007Q1 - 24 6 None Unempl. 1988M1 -2013M2 302 5 (2 2 1) (0 1 1) Retail trade Faiz Alsuhail/ Statistics Finland

How the study was made? For the sake of comparability we use the same seasonal adjustment settings as in Demetra. This includes: Transformation: Logarithm/ none; ARIMA model and mean correction; Trading day correction, correction for national holidays, correction for Easter effect Outliers, including type and timing; We are interested in the quarter-to-quarter (month-to-month) growth rates of the SA series. For the unemployment rate we are interested in the levels. Faiz Alsuhail/ Statistics Finland

Are the quarter-to-quarter growth rates (SA) different? GNP. Faiz Alsuhail/ Statistics Finland

Are the quarter-to-quarter growth rates (SA) different? LCI. Faiz Alsuhail/ Statistics Finland

Are the month-to-month growth rates (SA) different Are the month-to-month growth rates (SA) different? Volume index for industrial output. Faiz Alsuhail/ Statistics Finland

Are the month-to-month growth rates (SA) different? Retail trade. Faiz Alsuhail/ Statistics Finland

Differences in growth rate Average of Differences in Growth Rates (SA), percentage points Median of Differences in Growth Rates (SA), percentage points GNP 0,19 0,11 LCI 0,34 0,30 VIIO 0,64 0,50 Retail Trade 0,82 0,65 Faiz Alsuhail/ Statistics Finland

Are the levels (T) different? Unemployment. Faiz Alsuhail/ Statistics Finland

Unemployment J-Demetra+ Demetra Outliers Yes, 5 outliers TC1993m2 value 1,2896  1.2893 TC1993m2 t-value 3,69 3.68 TC1995m1 value 0,9689  0.9679 TC1995m1 t-value 2,77 2.77 TC1995m7 value 1,4558  1.4550 TC1995m7 t-value 4,17 4.16 TC1996m3 value -1,6482 -1.6471 TC1996m3 t-value -4,72 -4.72 TC2000m2 value 1,4398  1.4397 TC2000m2 t-value 4,13 4.13 Faiz Alsuhail/ Statistics Finland

Conclusion This study suggests that if an NSI is using Tramo/Seats then: Switching from Demetra to J-Demetra+ will cause revisions in the published figures. The differences seem to larger for time series with many trading day regressors. It might be good to consider: The timing of the switch. To which extend is it worthwhile to keep the same model settings as in Demetra. Communicating the change to the users. Faiz Alsuhail/ Statistics Finland