Statistical Office of the Republic of Slovenia (SURS)

Short-term Business Statistics (STS)

Litostrojska cesta 54, 1000 Ljubljana, Slovenia

The indices on turnover in wholesale, retail trade and sale and repair of motor vehicles are suitable for short-term observations and analyses of economic situation. The main objective of the monthly estimation of turnover in trade is early detection of changes in economic development. The indices of volume of sales represent the changes in the value of turnover in constant prices.

NACE rev.2.

The enterprises in Statistical Business register (SBR) whose principal activity is classified in section G - Wholesale and retail trade and repair of motor vehicles and motorcycles of NACE Rev.2. 

Turnover comprises the totals invoiced by the observation unit during the reference period for sales of goods and services. Sale on loan is included in the month of invoice. Turnover includes all other charges (transport, packaging, etc.), passed on to the customer, even if these charges are listed separately in the invoice. Reduction in prices and rebates as well as the value of returned packing must be deducted. Turnover excludes VAT, excise duties and other similar taxes, rentals, sales of fixed assets, other operating income, financial income and other extraordinary income. The volume of sales represents the changes in the value of turnover in constant prices. There is no inconsistency to EU definition (according to Regulation (EU) 2019/2152 of the European Parliament and of the Council, Commission Implementing Regulation (EU) 2020/1197 and current methodological guidelines).

The observation unit is the enterprise, whose principal activity is classified under NACE Rev. 2 section G - Wholesale and retail trade and repair of motor vehicles and motorcycles. Data refer to the whole enterprise, including eventual secondary activity not related to trade. In addition, there are some enterprises in Slovenia whose principal activity is not part of the observed activities or their secondary activity is part of other observed activities, but generate a significant share of their turnover with one of these activities as their secondary activity. Therefore, we survey also such major enterprises (if their principal activity is covered under trade or services surveys), but we take into account only the data related to the observed activity.

The survey frame is based on Statistical Business Register. Observation units are the enterprises whose principal activity is classified into trade (Section G of NACE Rev. 2). Data refer to the whole enterprise, including eventual secondary activity not related to above mentioned activity. In addition, there are several enterprises in Slovenia whose principal activity is not part of the observed activities or their secondary activity is part of other observed activities, but generate a significant share of their turnover with one of these activities as their secondary activity. Due to this fact our results were adjusted by having added the corresponding part of trade activities of these enterprises. The criteria for units whose principal activity is classified into trade to be included in the survey is a yearly turnover of the enterprise and number of persons in paid employment. For newly selected units yearly turnover is more than EUR 300,000, a yearly turnover is more than EUR 150,000 and more than 2 persons in paid employment or more than 5 persons in paid employment. For units from previous year the above mentioned turnover criteria lower from EUR 300,000 to EUR 180,000 and from EUR 150,000 to EUR 90,000.

Turnover from sale covers only activities carried out by the enterprises registered on the national territory of the Republic of Slovenia.

From January 2000 onwards. Until the end of 2022, the length of the series is 276 months. Only volume turnover for G46 is available from January 2010 onwards. The lenght of these series is 156 months

2015

Index.

A month.

Regulation (EU) 2019/2152 of the European Parliament and of the Council, Commission Implementing Regulation (EU) 2020/1197. The National Statistics Act (OJ RS No. 45/95 and No. 9/01) and the Annual Programme of Statistical Surveys.

For G45 and G47 we do not send data directly to any other data producing agency except to Eurostat. For G46 SURS has agreement with UMAR (Institute of Macroeconomic Analysis and Development of Republic of Slovenia) for distribution of more detailed but still aggregated data.

Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics and repealing Regulation (EC, Euratom) No 1101/2008 of the European Parliament and of the Council on the transmission of data subject to statistical confidentiality to the Statistical Office of the European Communities, Council Regulation (EC) No 322/97 on Community Statistics, and Council Decision 89/382/EEC, Euratom establishing a Committee on the Statistical Programmes of the European Communities. The National Statistics Act (OJ RS No. 45/95 and No. 9/01).

Only the aggregated data are published from which the single data values can not be recognized. 

The release calendar with advanced release dates for all months of the reference year is published at the beginnig of the year on a website of the Statistical Office. The release calendar is publicly accessible.

Release calendar 

The most important information channel for data publication is the SURS's website. The SURS occasionally publishes electronic and printed publications. Data are available free of charge, except those data that are prepared on users’ request. Statistical data and information are always published at 10.30 am. All releases are announced at the SURS’s website in Release calendar. None of the users have a prior access to the data on trade. The simultaneous issue of the First Release is ensured. Same release policy applied to national release is applied to transmissions to Eurostat.

Monthly.

The data on G45 and G47 turnover are published monthly at the SURS’s website according to release calendar, in the first release “Turnover in retail trade”. With releases in the First Release series SURS disseminates first and most important findings of statistical surveys.

There is no periodical printed publications. SURS on occasions publishes publications, which can contain data on trade.

SiStat database (Trade and services).

In Slovenia, the entities entitled to obtain the statistically protected microdata include the registered research institutions, registered researchers, and the researchers of government offices. Basic instructions concerning the access and the use of statistically protected microdata are availabe on the SURS's website.

The data on trade turnover are transmitted to Eurostat according to STS Requirements under NACE Rev. 2 in order to compile EU aggregates and for the dissemination of national data in Eurostat’s dissemination database. The data on trade turnover are published also in the publication of Slovenian Chamber of Commerce.

Methodological explanations for trade turnover data are availabe on the SURS's website:  Theme: Trade and Services, Sub-theme: Trade; Turnover in distributive trade, Slovenia. There is a link to the methodological explanations repository in the electronic version of the First Relase and an automatic link in the SiStat database for the data set on trade turnover indices.

Quality reports are available at the SURS's website: Theme: Trade and Services, Sub-theme: Trade; Turnover in distributive trade, Slovenia. 

Quality documentation can be reached on SURS's website.

Indices on trade turnover are produced in compliance with methodological requirements and standards.

The key users of the data are public sector (Slovenian Institute of Macroeconomic Analysis and Development, Ministry of Economic Development and Technology, National Bank of Slovenia, National Chamber of Commerce), business entities whose activity is trade, educational and research institutions (University of Ljubljana, School of Economics and Business, EIPF, d.o.o.), media, foreign institutions (Eurostat, ECB) and national accounts.

The principle users are asked about their needs, wishes, and interests at the regular meetings, i.e. at the Business Statistics Advisory Commitee.

Statistical Office of the Republic of Slovenia (SURS) also measured general user satisfaction for the last time in 2021. Respondents assessed general satisfaction with SURS with the average score of 8.3 (on a scale from 1 – disagree completely to 10 – agree completely).

All statistics that are required by current EU legislation are available.

The over-coverage rate for 2022 is on average 0.5% for G45, 0.5% for G46 and 1.0% for G47. The under-coverage bias for 2021 is 2.6 for G45, 0.8 for G46 and 0.0 for G47 (SBR data for 2022 is not available yet) and is a result of the procedure used for the determination of frame population. It shows the difference between annual indices calculated from data of the Statistical Business Register (SBR) taking into account the entire target population and annual indices calculated from SBR data taking into account the survey population (coverage).

The average weighted unit response rate of the first release (t+30) for 2022 is 40.5% for G45, 38.4% for G46 (data for this activity are not published yet at that point, but at t+60) and 79.9% for G47. Variable for calculation of weighted unit response rate is annual turnover of the unit. A significant number of units constituting the category of unit non-response rate of the first release represent units for which the data are obtained from VAT database. These data are available to SURS approximately 45 days (t+45) after the reference period.

Because of two different sources, we can talk also about field unit response rate and response rate for units whose data we get from the administrative source. Field unit response rate for 2022 is 99.1% for G45, 98.1% for G46 and 99.6% for G47 (already one month after the end of the observation period the response rate for the field units is very high; if a unit does not report in time, we obtain data from it later on). The average response rate for units whose data we get from the administrative source is 91.3% for G45, 95.7% for G46 and 89.2% for G47.

The average imputation rate for the missing data and data obtained from administrative sources, which were at the stage of automated data editing marked as to prominent, is 10.3% for G45, 4.2% for G46 and 9.4% for G47.

The survey is not conducted on the basis of a random sample, therefore there is no sampling error.

Coverage errors

The over-coverage rate in 2022 is on average 0.5% for G45, 0.5% for G46 and 1.0% for G47. The units which represent over-coverage can be divided into ineligible units (we include them in coverage because of lack of information, but do not belong there) and units which were at the inclusion in coverage eligible, but became ineligible at the time of survey implementation as a result of demographic changes.

Under-coverage is a result of the procedure used for the determination of frame population. The survey frame is based on Statistical Business Register (SBR). Observation units are chosen at the end of the year for the next year on the basis of threshold coverage. We use the cut-off sample selection; the whole cut-off procedure is carried out in two steps. In the first step the units included in the survey are determined and then in the second step the units for which the data will still be obtained by questionnaire are selected. The enterprise is included in the survey if it fulfils one of the following criteria:

• have annual turnover higher than EUR 180,000 for the units from previous coverage or EUR 300,000 for newly selected units,
• have annual turnover higher than EUR 90,000 and more than 2 persons in paid employment for the units from previous coverage or EUR 150,000 and more than 2 persons in paid employment for newly selected units,
• the enterprise has more than 5 persons in paid employment.

The units for which the data will be obtained by questionnaire are determined on the basis of the reported turnover from the previous year. We first sort the units by descending turnover. Then we select a sufficient number of units (largest units) from the beginning of the list so that they exceed the defined share (approximately 60%) of turnover in the total turnover of units covered in the selected activity group. Every year we calculate the under-coverage bias on the basis of SBR data. We compare the year-on-year indices (average of the T year compared to average of the T-1 year). The difference between annual index calculated from data on SBR taking into account the entire population and annual index taking into account the survey population (coverage) for 2021 is 2.6 for G45, 0.8 for G46 and 0.0 for G47 (SBR data for 2022 is not available yet).

Multiple listings: We haven’t detected yet that the enterprises are present more than once in the frame, since the trade surveys are based on SBR.

Measurement errors

We collect data on turnover directly from the most important observation units with the monthly questionnaire. The reporting units can choose only the electronic reporting of data. The paper version is not on availability. The most important reporting units must send full-filed questionnaires to the Statistical Office by the 8th day in the current month with the data for the previous month. The main sources of data on turnover for other units included in the survey are the data which are reported by enterprises to the Financial Administration of the Republic of Slovenia for the purpose of value added tax return (DDV-O form).

There are some methodological differences in the definition of turnover between short-term statistics and the VAT database. The comparison of the index series calculated from two different sources (field units and a combination of field and VAT units) in feasibility studies indicate that VAT data can be used as the main source of turnover. For field units for which data are obtained through a questionnaire, we have the data from the VAT database, so that this (sub)population can be compared to the consistency of data from two different sources. The average rate of the coherence in terms of absolute difference of indices for 2022 is 0.3 for G45, 0.2 for G46 and 0.1 for G47. If the respondents give the erroneous data on monthly turnover they always have the opportunity to report the correct value later on. If the erroneous data are significant then we detect them through the logical controls, otherwise we detect them by coincidence. Fundamentally there is no interviewer impact in the trade surveys since the data are gathered on electronic questionnaires. The impact of the interviewer is present just in the case when the unit is contacted by phone because of non-response, but we do not evaluate this kind of errors.

Non-response errors

Unit response rate: The average weighted unit response rate of the first release (t+30) for 2022 is 40.5% for G45, 38.4% for G46 (data for this activity are not published yet at that point, but at t+60) and 79.9% for G47. Variable for calculation of weighted unit response rate is annual turnover of the unit. A significant number of units constituting the category of unit non-response rate of the first release represent units for which the data are obtained from VAT database. These data are available to SURS approximately 45 days (t+45) after the reference period. For these units we estimate monthly turnover for the first release (t+30). With the consideration of estimated monthly VAT data in the category of unit response the average weighted unit response rate is 96.6% for G45, 97.4% for G46 and 98.6% for G47. A number of units constituting the category of VAT unit non-response report VAT data on a quarterly, rather than on a monthly basis. For these units we estimate monthly income for the three months of the quarter at the end of the quarter. With the consideration of estimated monthly VAT data in the category of unit response the average weighted unit response rate is 99.6% for G45, 99.3% for G46 and 99.9% for G47.

Because of two different sources, we can talk also about field unit response rate and response rate for units whose data we get from the administrative source. Field unit response rate for 2022 is 99.1% for G45, 98.1% for G46 and 99.6% for G47 (already one month after the end of the observation period the response rate for the field units is very high; if a unit does not report in time, we obtain data from it later on). The average response rate for units whose data we get from the administrative source is 91.3% for G45, 95.7% for G46 and 89.2% for G47.

Item response rate: We monitor only one variable (turnover), so the item response rate is equal to the unit response rate in general.

Data processing errors

For field units for which data are obtained through a questionnaire the control takes place interactively so we do not have the accurate records of the editing rate, but only an estimate. For 2022 we estimate that we checked the data for approximately 15% of field units and actually corrected data for approximately 2.0% of units. For the units for which the data on turnover are obtained from VAT database the automatic data editing is performed, where all adjustments take place through computer applications. In the survey we impute (fill in) the missing data and data obtained from administrative sources, which were at the stage of automated data editing marked as too prominent. The average imputation rate is 10.3% for G45, 4.2% for G46 and 9.4% for G47. Impact of coding is not relevant because coding of response data is not used.

Model errors

Model errors are not relevant because no specific models are used in estimation.

The publication schedule is fixed and announced in advance. The First Release is published 30 days after the end of the reference month. Data are at the same time also loaded into the SiStat database on SURS's website, created for user friendly free of charge data manipulation. Data are published as provisional and they are revised monthly as the response rate increases. Data are considered as provisional 10 months after the reference period.

There is no time lag between the actual delivery of the data and the target date when it should have been delivered.

The survey is based on Regulation (EU) 2019/2152 of the European Parliament and of the Council, Commission Implementing Regulation (EU) 2020/1197 and current methodological guidelines. These documents take into account all the European countries which transmit their data about trade to Eurostat. This ensures good comparability of data between countries and good quality of aggregates calculated from these data at the EU level. Because of various specialties, for instance different methods of data collection (field surveys or usage of administrative resources) and different ways of calculating indices, there may be differences that affect the comparability of data.

Time series of indices calculated according to the current methodology are available from January 2000 onwards. Until the end of 2022, the length of the series is 276 months. Only volume turnover for G46 is available from January 2010 onwards. The lenght of these series is 156 months. On 1 January 2008 all Member States of the European Union launched a new classification of activities of business entities NACE Rev. 2, which replaced the existing NACE Rev.1.1. In Slovenia the national version of the standard classification, called SKD 2008, came into force. It includes the entire European classification of activities, but also adds national divisions. We started publishing trade statistics according to the SKD 2008 in 2009 for the reference periods in 2009. Due to changes in the classification of activities, the time series of statistical data had to be recalculated (backcasting) according to the new classification (from 2000 to 2008). A significant difference in the monitoring of turnover in retail trade under the current SKD 2008 is the inclusion of sales of motor fuels in the retail trade aggregate. Sale of motor fuels was, according to SKD 2002, monitored within an aggregate sale and maintenance of motor vehicles and retail sale of automotive fuel.

The results of this survey for G47 are comparable with the results of the monthly survey Business Tendency in Retail trade (PT-TRG). The purpose of the qualitative Survey on Business Tendency in Retail Trade (PT-TRG) is to obtain monthly information about current major economic indicators in retail trade and to evaluate their movement in the ensuing months. The results of the survey are the basis for evaluating the confidence indicator in retail trade and for the sentiment indicator. The observation units in this survey are enterprises of which the main activity in the Business Register of Slovenia is classified according to the NACE Rev. 2 as retail trade.

The comparison between the confidence indicator and indices of retail trade turnover for previous years showed good correlation, but has changed in recent years. For the period from January 2012 to December 2022 it is 0.26. The change can be attributed to uncertainties regarding COVID-19 period but also to various trends on the market, some changes in business demography and structural changes which all coincided in the second half of 2021. The trend also continued in 2022 and was prolonged by the war in Ukraine and inflation.

Due to the different nature of the data, differences between the results of one or another survey are not indicated, but the results are shown only in graphic form.

Notes: The surveys differ in methodologies, since PT-TRG is a qualitative survey aiming at monitoring the opinions of managers on certain economic indicators within the enterprise, while the TRG/M survey is a quantitative survey aimed at monitoring value variables.

We can also compare data with the data of the annual Trade (TRG/L) survey. The data from the monthly survey are recalculated to the annual indices and the nominal absolute data from the annual survey are recalculated to the indicies.

Notes: The surveys differ in methodologies, since annual survey only covers the turnover on sale of trading goods (without the sale of own products and services) on the domestic market. Monthly surveys cover total turnover (including from the secondary non-trade activities) on both domestic and foreign market.

Data are internally coherent. Each level of aggregated data is compiled directly from micro data.

In 2022 the number of working hours spent for the selection of enterprises in the sample, preparing the address book, designing, printing and distribution of notification letters, data capture, data entry and data verification, calculation of the results and their publication, etc. was around 3,000. Number of reporting units that submitted the data by questionnaire for one month was on average 105 for G45, 171 for G46 and 232 for G47, annual number of questionnaires per unit was 12.

Types of data revisions in relation to planning:

a) Planned data revision is subject to the following reasons:

Due to the needs of users for timely information, data are published that meet the criteria of the quality of official statistical data, but do not meet the quality that can be met with additional statistical procedures. Final data are based on more complete answers about the phenomenon and/or analyses and are published later on;

Seasonal adjustment and/or elimination of calendar effects;

Change in methodology and classifications;

Due to change of base year.

b) Unplanned data revision is not part of the regular statistical process. It appears due to unpredictable changes in the methodology, unpredictable emergence of new and better data, unpredictable changes regarding reporting units that transmit their data to the Office, unpredictable obstacles in data processing and publishing, and errors in data processing and publishing (e.g. a key unit corrects its data for the past few months, an unpredictable change in the administrative data source).

Types of data revisions in relation to time of implementation:

a) Regular revisions: inclusion of a more complete/additional data source or a change in the data source, seasonal adjustment and/or elimination of calendar effects;

b) Occasional revisions are a consequence of including a new/more complete/additional data source that becomes the standard in later data releases or a consequence of an unpredictable obstacle in data processing and publishing, and change in methodology.

Types of data revisions in relation to the purpose:

a) Inclusion of a more complete/additional data source or a change in the data source;
b) Seasonal adjustment and/or elimination of calendar effects;
c) Transition to a new base period;
d) Improvement of methodology due to a change in the statistical method or a change in classifications, concepts and definitions;

e) Elimination or errors

More information about data revision are published in the methodological explanation on revision of statistical data.

Same revision policy is applied nationally and in transmissions to Eurostat. Regular analysis between provisional and final data are carried out for the national STS data sent to Eurostat (published in annual quality report). In Quality Reports the difference between annual indices calculated from the retail trade survey (for administrative and field units together) and annual indices calculated from the SBS data is calculated. For field units for which data are obtained through a questionnaire, we have (with rare exceptions) the data from the VAT database, so that this (sub)population can be compared to the consistency of data from two different sources. Planned revisions are part of statistical process and are in correspondence with release calendar (see the concept 8). 

Data revision of trade indices is planned. Due to the requirements of the Regulation (EU) 2019/2152 of the European Parliament and of the Council, Commission Implementing Regulation (EU) 2020/1197 and needs of users for timely information, data are published that meet the criteria of the quality of official statistical data, but do not meet the quality that can be met with additional statistical procedures. Regular revisions of trade data include a more complete data source (because of late unit reasponse or updates of data for the previous months or inclusion of VAT data)  and change of data due to seasonal adjustment and elimination of calendar effects. Data on retail trade for the last 10 months are provisional. With each release on a monthly basis the data for the last 10 months can be corrected and supplemented with new data. Due to change of the base year a revision is planned every five years. Data was revised due to change of the base year in 2018 (base year = 2015). Changes and revisions of methodology are announced in advance in our publications. Before that the Advisory Committee discusses the proposed methodological changes. In a revision due to an error in data, a new edition with explanatory notes is published as soon as possible. Due to the changes in data of some enterprises, there was a unplanned revision in october 2021, regarding the G46 data for the years 2019, 2020 and part of 2021.

The Mean Revision (MR) and Mean Absolute Revision (MAR):

Growth rate for calendar adjusted data (year-on-year) for reference period 2019 - 2021 (36 months):

There is no impact of the base year change since we use a chain linking in calculation of index.

In order to collect data on trade turnover a combination of two sources of data is used: main sources of data on turnover are the data reported by enterprises to the Financial Administration for the purpose of value added tax return. In this way about 90% of the data in terms of number of units are obtained. Administrative data being reported by enterprises to the Financial Administration of the Republic of Slovenia for the value added tax purpose (DDV-O forms). These data are available to the Statistical Office approximately 45 days after the reference period. In order to provide results that are close enough to the statistical definition of the observed phenomenon and to stay in touch with the most important reporting units, we collect data on turnover directly from about 10% of the largest units with the monthly questionnaire. Reporting units are obliged to deliver the requested data for the previous month to the Statistical Office no later than the 8th of the month. The source for these data is the accounting documentation of enterprises and only exceptionally their estimates.

Monthly.

For most of the units VAT declarations (DDV-O form) are used. Data for the largest units are collected by electronic questionnaire. Actions to speed up or increase the rate of response: at the first one postal reminder is used, which is sent by mail usually 3 days after the time given for the replies. There is also a telephone follow up.

Plausibility checks performed at micro level: if the field unit's turnover for the reference month is not comparable to its turnover for the previous month or the same month of the previous year (lower or higher than a defined threshold) than the unit is contacted by phone in order to clarify the deviation. The data obtained from administrative sources, which were at the stage of automated data editing marked as too prominent are imputed. Plausibility checks performed at macro level: we compare the indices calculated for the units for which the data are obtained from administrative data to indices calculated for the field units. Comparison of the index on aggregate level for the reference month with the indices of the same months of the previous years is also made on aggregate level. A kind of detection question on demographic change of the unit is added on the questionnaire (did the unit operate in the reference month and if not why not). There is also the remark field for any explanation of the unit (especially for the case of very low or high turnover). There is also possibility to report the corrections of turnover for the previous months. The objective of all kind of the detection questions on the field questionnaire is to get more relevant information on the data important for the survey. For the field units we collect the contact data of the person which fulfills the questionnaire and in the case of uncertainties in the data we make an inquiry calls. The information and explanation of deviated data gathered directly from the enterprises are considered as very precious. Sometimes the information from enterprises explain the unusual movement or development of data series. Sometimes the enterprises do not inform us about the demographic change automatically but we reach the information through inquiry call. In the case of uncertainties in administrative data no direct contact to the units exists.

Within the survey, only one variable is measured – turnover. Turnover is also the only one variable which is estimated within the survey in the case of non-response. In t+30 VAT data are not available yet and the VAT units are considered as non-respondents in the first month. Historical trend method is used for the non-respondents for which the data for the previous month exist. Interpolation of quarterly data is used in the last month of each of the quarters for the units for which only the quarterly VAT data are available. The quarterly data are broken down to the monthly data by using the nearest neighbour principle. And average value (Mean-Value Method) within the stratum is used for the units for which one of the afore mentioned methods could not be used. No weights are used for aggregated data. The each level of aggregated data is compiled directly from micro data (the indices on the aggregated level are only self weighted). A simple value index is compiled. The only index which is calculated directly from the turnover data is the month-to-month change. After calculating this index, the »fixed base year index« is calculated by multiplying the base year index from the previous month with the month-to-month change. All the other indices are calculated by chaining the time series of the base year indices.

Seasonal adjustment is the main part of time series analysis. With seasonal adjustment, seasonal effects and calendar effects are removed, if they are significant. In this way, the data are simplified so that they can be more appropriately interpreted, because seasonal fluctuations can blur other important movements.

JDemetra+ 2.2.0 software, TRAMO/SEATS method, is used for seasonal adjustment. In seasonal adjustment the guidelines of the European Statistical System are also taken into account. For each new release of unadjusted data, the parameters of the models are re-estimated; the models mostly remain the same, but sometimes they are changed (e.g. an outlier at the end of a time series is added). Major changes of the models usually occur every few years. For each new release of unadjusted data, the whole seasonally adjusted time series are revised. Models are revised in detail about once a year, taking into account the period available at that time. If the model is changed, we strive to minimize the changes. In the period between detailed revisions, the model is changed only if statistics show that it is no longer good enough.

Selection of the model is manual, but some automatic tests are used for help (test for transformation, automatic detection of outliers, automatic selection of ARIMA model). All 10 transmited time series have multiplicative decomposition. Critical value for outlier detection depends on the time series, usually it is between 2.5 and 3.5. Most of the time series don't have the possibility of outlier detection. Filter length is automatically chosen. All the time series are seasonally adjusted directly, therefore seasonally adjusted data of an aggregate are not composed of seasonally adjusted data of its components. When models are selected or changed, connection between an aggregate and its components is taken into account (similar time series have similar models …). Residual seasonality is checked when the model is selected. Afterwards, residual seasonality diagnostics are taken into account.

The results of seasonal adjustment depend on the software, method and model used. In 2016 we started using JDemetra+ instead of Demetra+ software. Time series models were carefully revised and, if necessary, changed. The entire period was taken into account. Due to software replacement and changed models there were no longer revisions in seasonal adjustment results.

No comment.