External trade in goods

Name: External trade in goods
Topic: External economy

Division for External Trade Statistics

Statistics on imports and exports of goods are published at country level. Export figures are also given by county of production.

Monthly figures are released on the 15th of the month after the observation period (the previous month), or the first subsequent working day. With every new release, all the previous monthly figures are updated in every publication.

For more information about Revisions, please see Accuracy and Reliability.

Reports are sent to the EU statistical office (Eurostat), Organization for Economic Cooperation and Development (OECD), World Trade Organization (WTO), International Monetary Fund (IMF) and the United Nations (UN).

Non-revised and revised micro data are stored in accordance with Statistics Norway's guidelines for storing computer files (DataDok).

The purpose of the external trade statistics is to provide information about the commodity flows between Norway and other countries. Exports and imports are important economic indicators both in describing structural changes and in monitoring the economic trends. The monthly statistics of external trade in goods have been published since 1913 (value figures were published for the first time in 1866).

With regard to essential changes in the statistics, the commodity classification in line with the Harmonised System that was introduced in 1988 entailed a comprehensive restructuring of the distribution at a detailed level in relation to the earlier CCCN nomenclature (the Customs Cooperation Council Nomenclature).

The statistics are mainly financed from State assignments, but also rely to some extent on market income.

The external trade statistics are a part of the national and international statistical system, and are used in the compilation of the national accounts and the balance of payment statistics in Statistics Norway. The statistics are aimed at covering the business sector’s need for data in this area for their planning of production, marketing and sales.

The statistics are also intended for prognoses and analyses performed by public authorities, research institutions and private organisations. International organisations such as the UN´s statistical office, the Organization for Economic Cooperation and Development (OECD), the statistical office of the EU (Eurostat) and others make use of the information provided in the external trade statistics.

No external users have access to the statistics and analyses before they are published and accessible simultaneously for all users on ssb.no at 8 am. Prior to this, a minimum of three months' advance notice is given in the Statistics Release Calendar. This is one of Statistics Norway’s key principles for ensuring that all users are treated equally.

The figures for external trade statistics are an important part of the quarterly balance of payments and national accounts and the annual national accounts.

The balance of payments also collects estimates of trade in goods which are transported directly from abroad to the Norwegian continental shelf, imports of fuel and other oil products to Norwegian ships, aircraft and units abroad. Moreover the external account interprets some of the goods from the external trade statistics in goods as a service, such as computer games, software and licenses.

In the publication of the figures for external trade we give figures for crude oil, natural gas and condensates, while the balance of payments use the collective term “crude oil and natural gas”. Terms in foreign accounts also encompass items such as naphtha, propane and butane, while the external trade statistics do not.

In addition to the statistics mentioned, external trade figures are included in many of the other statistics published by Statistics Norway, for example:

• Exports of salmon, weekly figures of salmon exports are published in addition to the regular monthly trade figures
• Indices of volume and price of external trade in goods mainly use external trade data as the basis for calculating figures. In addition, the total number of imports and exports, as well as fish and some selected commodity numbers for waste paper and scrap metal, are used as basic data in the producer price index and the price index of first-hand domestic sales
• External trade data are also used as basic data for statistics on fishery and emissions of greenhouse gases.

Mirror statistics
Mirror statistics, in the context of external trade statistics, refer to a comparison between the statistics of two or more countries: exports of goods from one country should in principle equal the imports of the same goods in the partner country.

However, discrepancies are often identified in these figures for several reasons. Although there are international guidelines for the production of external trade statistics, there may still be differences in what the countries include. An example is differences in the usage of statistical trade systems, general or special trade, which causes delimitations to the statistics content. Countries may also have different methods for data collecting and processing, which can create distortions. Furthermore, some countries collect data from customs authorities and others directly from companies.

Basically, there should be a certain difference in value between the countries, as the exporting country provides the FOB-value while the importing country in addition to this adds costs for insurance and freight all the way to the location of delivery, the CIF-value (see Definitions). Some countries collect data from customs authorities, others directly from companies. A common source of error between countries is caused by different information about which is the country of destination. Different threshold-values for inclusion of data can also create distortions, f.ex. in Norwegian external trade statistics, declarations with value less than NOK 1 000 are excluded from the import statistics. Other countries may have other delimitations. Different practise of confidentiality and demands for confidentiality, inaccurate definitions of certain commodities as well as conscious and unconscious misuse of codes can lead to use of different classification codes between countries.

From a Norwegian point of view, there is a need for comparing external trade data with the corresponding statistics of our main trading partners, principally European countries. It is in this context important to have knowledge of the rules practiced by the individual partners.

The Statistical Office of the European Union, Eurostat, publishes external trade data for EU countries, totally for the Union and separate for each member country. This statistics are produced according to EU regulations, which defines external trade as trade between the European Union trade and third countries. As a consequence of this, exports to countries in the EU from a non EU-member state are recorded as imported to EU in the first country where the goods cross the borders of the European Union. When forwarding the goods to the country of destination, the good is recorded as an item dispatched from the EU-country that first received the goods. Information about the country of origin of the commodity is not included in the records. Several of the EU-countries publishes in addition to the data presented in the Eurostat database, Comext, separately external trade statistics produced after national methodology. For information about this, consult Eurostat's annual report: Quality report on International trade statistics.

The Statistics Act of June 21, 2019 nr. 32, §10 and §20

EU regulations incorporated into the EEA agreement comprise contractual obligations pursuant toto EC regulations onon data collection and distribution of data to the EU’s statistical office, Eurostat. The current regulations are as follows:

Commission Regulation (EU) No. 92/2010 of 2 February 2010 implementing Regulation (EC) No. 471/2009 of the European Parliament and of the Council on Community statistics relating to external trade with non-member countries, as regards data exchange between customs authorities and national statistical authorities, compilation of statistics and quality assessment

Commission Regulation (EU) No. 113/2010 of 9 February 2010 implementing Regulation (EC) No. 471/2009 of the European Parliament and of the Council on Community statistics relating to external trade with non-member countries, as regards trade coverage, definition of the data, compilation of statistics on trade by business characteristics and by invoicing currency, and specific goods or movements

Commission Regulation (EU) No. 1106/2012 of 27 November 2012 implementing Regulation (EC) No. 471/2009 of the European Parliament and of the Council on Community statistics relating to external trade with non-member countries, as regards the update of the nomenclature of countries and territories

Statistics area and the economic territory
The scope of foreign trade statistics is based on international guidelines from the United Nations Statistical Commission.

The general recommendation is that the statistics should include all goods which add to or subtract from the stock of material resources of a country by entering (imports) or leaving (exports) its economic territory. In addition to the geographical dimension, a country's economic territory also includes ships, aircraft and oil platforms owned by national corporations. These are goods that are not necessarily located within the geographic area, but are still viewed as part of the material resources of a country. For these goods external trade is measured by the change of economic ownership.

There are two different statistical trade systems - the general trade system and the special trade system. UN recommends using the general trade system where a country's statistical area coincides with its economic territory. This implies that goods are recorded at the time they enter or leave the economic territory. The special trade system is used when the statistical area covers only a part of the economic territory. Statistics Norway follows, with a few exceptions, the general trade system.

The Norwegian economic territory consists of mainland Norway, Svalbard, Jan Mayen, Norwegian dependencies, territorial waters, including the airspace above these, the economic zone, the Norwegian part of the continental shelf, territorial enclaves abroad (embassies, consulates, military bases, research stations) and Norwegian ships, aircraft and oil platforms.

External trade statistics is mainly based on information drawn from customs declarations. However, the responsible area of the customs authorities only covers mainland Norway and its territorial waters. As a supplement, data on important trade in goods to and from the remaining areas of the economic territory are collected directly from respondents and registers.

The following are defined as statistical area:

Mainland Norway and its territorial waters
Customs declarations contain information on physical flows of goods crossing the customs border. Imports and export that are exempted from ordinary declaration are referred to in the Customs Act §§ 4-10 and 4-11. According to the recommendations from the UN, some flows of goods should be excluded from the statistics. This includes trade in Norwegian goods between the customs area and other parts of the Norwegian economic territory. To the extent possible, trade in foreign goods to and from foreign enclaves, ships, aircraft and oil platforms located in Norway are omitted. To distinguish this type of transactions customs procedure codes from the declarations are used. In general, trade transactions are recorded at the time when the goods enter or leave the customs territory. However there are some exceptions, cf. the treatment of goods in customs warehouses.

There are four different types of customs warehouses. Goods waiting on declearance,can be stored in either a general or a central warehouse. For imports the time of recording is set when the goods are taken out of the warehouse and not when they arrived in the economic territory. This means that the statistical processing of these data follows the special trade principle. However, in practice the time difference between storage and withdrawals from warehouses is usually so small that the statistical processing is considered to be approximately equal to the general trade principle. For the remaining two types of customs warehouse; duty free and storage for further processing, the statistical treatment is according to the general trade system.

Customs declarations of exports and imports of electric current do not provide enough information for statistical purposes, and data is therefore obtained directly from the respondent.

Svalbard and Jan Mayen
Import transported through the Norwegian mainland to Svalbard and Jan Mayen should be declared by customs. Data on exports of coal transported directly from Svalbard are collected directly from the respondents.

The Norwegian part of the continental shelf
Imports via the Norwegian mainland to Norwegian owned installations and vessels that operate on the continental shelf should be declared by customs. Exports of crude oil and natural gas delivered directly from the continental shelf are based on data from the Norwegian Petroleum Directorate and the different operators.

Norwegian owned ships, aircraft and oil platforms
Exports and imports of ships, aircraft and oil platforms are based on information from Norwegian ship registers as well as from different respondents. Export of Norwegian goods (excluding bunkers) to foreign ships and aircraft in Norwegian ports and airports are covered by customs declarations.

Data on exports of fish caught by Norwegian fishing vessels outside the customs area and landed abroad is obtained from the Directorate of Fisheries.

In accordance with the UN recommendations the following trade in goods are not included in the statistics:

• Goods in transit, ie goods attending the Norwegian economic territory to be directly transported to a recipient outside this area
• Goods traded through intermediate trade, ie goods transported between two foreign countries without entering the Norwegian economic territory
• Goods physically crossing the country border, but defined as a service (including newspaper subscriptions, accompanied baggage, household goods)
• Content delivered electronically (including e-books, downloadable games, etc.), defined purely as a service
• Issued banknotes, securities and monetary gold
• Samples, gifts and promotional material etc. with a low value
• Goods for repair or return after repair
• Goods in unaltered condition in return to vendor (claims etc.) and waste containers, replacement deliveries and free repairs (warranty)
• Items for temporary use, that are returned within 12 months (exhibition, demonstration, scientific research, loan, professional equipment, transport vehicles in international traffic, containers, equipment in use for the press, radio and television equipment, etc.)
• Change in ownership of non-financial assets
• Goods that are damaged on the journey to and from the Norwegian economic territory
• Satellites equipment moved to and launched in other countries without any change of ownership
• Goods functioning as means of transport (ship, aircraft, packaging, etc.

Furthermore goods totalling less than NOK 1 000 are omitted, these goods amounts to less than one per cent of the import and about 0.04 per cent of the export. Imports of goods for less than NOK 200 and exports of goods less than NOK 5.000 (except from goods that are subject to taxes and restrictions) are not subject to declaration by the Customs.

The statistics are mainly based on administrative information obtained from TVINN, which is the Norwegian Customs’ electronic information system for the exchange of customs declarations between businesses and Norwegian Customs. Some data is, however, obtained from other sources, and more information on this can be found in the paragraph on data collection.

Sampling is not not relevant as external trade is a total count of the goods defined in the population.

Data collection

The vast majority of the data is, as mentioned, collected through the Customs’ TVINN register and transmitted electronically to Statistics Norway. However, some additional data, as shown below, is obtained separately.

Exports of crude oil and natural gas in gaseous state
Preliminary and final figures for exported volumes of crude oil and natural gas in gaseous state are collected through monthly reports from the operators and the Norwegian Petroleum Directorate. The prices, however, are obtained via various sources depending on whether they relate to crude oil or natural gas, and whether the data is being collected to compile preliminary or final figures.

From 2013 on, the development in the in gas prices from the producer price index (PPI) used to calculate the preliminary gas prices for foreign trade, while the level will be based on the value determined from the last registered price. PPI obtain figures for natural gas from different open sources for gas prices in the European market.

Preliminary oil prices are based on the reference price of Brent Blend for the current month.

For crude oil, the final prices: quarterly surveys from the licensees, as well as information from the Norwegian Petroleum Directorate, the Ministry of Petroleum and Energy and the operators.

For natural gas, final prices: a quarterly average price based on information from a selection of licensees and operators was used until 2019. From 2020 on we have monthly prices also for these figures.

Due to insufficient information about country of destination there are from 2013 on changes in the dissemination of the figures of exports of natural gas in gaseous state. From 2013, country of destination according to these figures will be based on the landing point of the pipelines with natural gas from the Norwegian continental shelf.

Exports of crude oil in pipelines to Great Britain
All exports of Norwegian crude oil that are transported by pipeline to Teeside are registered in the official statistics with the UK as the country of destination. The ownership is Norwegian until the oil is processed in Great Britain, but because a commodity is registered as exported when it physically has crossed the Norwegian border.
After this unstabilized crude oil has gone through a fractionation process, where it is separated into the components of stabilized crude oil, ethane, propane and butane, it is shipped onwards, thus giving a new country distribution. Crude oil has to be stabilized in order to be transported by ship.
Only those shipments regarded as genuine importation into Great Britain is regarded are included in the statistics of Great Britain. All other flows are regarded as being in transit, and therefore excluded.

The figures by country in Statbank table 12780 for “Crude oil, incl. oil by pipelines after processing in Great Britain” shows reallocated figures after country of destination for oil exports by pipe lines (to Great Britain) and by ship to all countries.
Only those shipments regarded as genuine importation into Great Britain is regarded are included in the statistics of Great Britain. All other flows are regarded as being in transit, and therefore excluded.
The fractionation process implies that some of the crude oil disappears, thereby different total figures.

Imports and exports of ships and oil platforms
The data collection for vessels is based on current information from the Norwegian ship registers the Norwegian Ordinary Ship Register (NOR) and the Norwegian International Ship Register (NIS).

When it comes to oil platforms, information is collected from the Norwegian ship registers, the Customs’ TVINN register and importer/exporter.

Imports and exports of electricity
Information is received from Statnett. The volume of imports and exports of electricity is measured by the net transfer of the power grid junction point towards foreign countries and is registered every hour. The value of this trade is calculated by multiplying the amount of electricity transferred per hour with Norpools market price for the current hour. Thereafter the figures are aggregated to a monthly level.

From October 2025, the method for calculating electricity has been changed from hourly to 15-minute intervals. This creates a discontinuity in the data, and users should be aware that the figures before and after this date are not directly comparable.

Imports and exports of aircraft
For some aircrafts, monthly reports are received from Statistics Sweden, while the rest are obtained through regular declarations in the TVINN system.

Exports of coal from Svalbard
Reports on the export of coal from Svalbard are received yearly from the exporter.

Exports of fish caught by Norwegian vessels outside the Norwegian customs border
In the period July 2010 - December 2023, the figures have been obtained from the Directorate of Fisheries (FDIR). Prior to and after that, the information was gathered through TVINN.

Lov-value shipments besides VOEC consignments
For commodity codes 99.60.1000 and 99.60.2000 cover customs-declared goods, and the information is obtained from customs declarations. The goods are divided into two commodity codes indicating whether the recipient/sender for imports/exports is a consumer or a business. Due to simplified declaration procedures, statistics for low-value shipments are less detailed. The trade is aggregated into total values, and the country of origin or destination is unspecified.

In the publication of May 15th 2025, commodity codes 26.21.1000 and 38.25.1000, which cover municipal waste, were expanded to include observations with value less than NOK 1.000. The change applies from 2023. This will result in a break in the time series for these commodities from 2022 to 2023.

Import of VOEC consignments
VOEC (VAT on E-Commerce) is a simplified scheme established for sellers and online marketplaces to charge and report VAT on business-to-consumer supplies of low value goods. Information about VOEC goods is collected from administrative data from the Norwegian Tax Administration. Each quarter, foreign VOEC-registered businesses must report and pay the VAT collected during the previous quarter. Corrections for earlier periods can also be included in the VAT returns, due to misreporting, returns of goods, or other reasons. The value of VOEC imports is calculated based on the VAT basis, adjusted for any corrections.

Due to quarterly reporting, monthly figures are estimated by dividing the quarterly amount across the relevant months. This reporting frequency means there is a time lag in the monthly statistics.

Due to insufficient information about volume, Statistics Norway has estimated the weight of VOEC goods based on private individuals' imports in the TVINN data. The calculation basis is limited to goods valued up to NOK 3,000, corresponding to the value limit in the VOEC scheme.

As with other low-value shipments, trade in VOEC goods is less detailed due to exemption from customs declaration.

Re-exports and returns of low-value shipments
Generally, exports of goods less than NOK 5,000 are not subject to declarations by the Customs. A large share of the low-value shipments that are re-exported will therefore not be declared in TVINN. Corrections reported in VAT returns to the Tax Administration for VOEC goods contain no identifiers to distinguish between misreporting, returns of goods, or other factors.

Data editing
Information from customs declarations is subject to statistical controls in the Customs declaration system (TVINN). The controls to check the estimated price, quantity and country were developed by the Customs in cooperation with Statistics Norway.

Data controls at Statistics Norway take place at different levels - some as soon as the data is loaded into the system (automatic recoding of variables). The manual part of the revision consists of different types of validity and probability tests. The former are controls of absolute errors, such as missing transport code, while the other type of test may check unlikely country codes or unreasonably high quantities for a specific type of good. The system aims to intercept the most serious errors in the data. Data collected from sources other than the Customs also undergo similar checks.

Estimations
External trade is based on a total count but some items are excluded, cf. section of population. There is no need for any statistical estimations of current figures, only aggregations, in order to enable us to publish in a timely manner.

The methods and routines used to produce seasonally adjusted figures are described in the chapter About seasonal adjustment.

The Statistics Act § 7 covers the disclosure of information.

The requirements for confidentiality in external trade of goods are practised so that data is suppressed only when the establishment/enterprise has a valid reason for confidential treatment and actively asks for it. Exporters/importers can demand suppression when there are 3 or fewer companies that import or export an item type to a given country, or when a company has a market share of over 90 per cent, or if two firms have a market share of over 95 per cent.This is called "passive confidentiality" and is in line with UN recommendations. This method is used by most countries..

Since Norway has a number of major manufacturers that are dominant in some product areas, some suppression of information exists on the export side. Between 5 and 10 per cent of mainland exports are affected by confidentiality. For imports, it is far less; around 1 per cent. Some commodity areas are more problematic with regard to dissemination, since suppression on highly aggregated levels must be carried out.

The six-digit HS-based commodity classification is subject to revisions due to technological development and also development in the commodity spectres of the international trade. These adjustments are normally made every five years. The most recent HS revision was implemented on 1 January 2012. Furthermore, national adjustments are made every year at the national eight-digit level. These revisions can create breaks in the time series for some commodity numbers. Detailed information about this is available in the yearly Commodity List External Trade. Correspondence tables between the different editions of the HS are available on the UN website.

With regard to the divisions according to the SITC, the current SITC-Rev.4 was introduced in January 2007, when it replaced the previous SITC-Rev.3 (1988-2006). The relation between the figures for the years before and after 2007 can be misleading for some commodity numbers. The degree of influence depends on the value of the commodity numbers that now have a different group affiliation. Correspondence tables between the different editions of the SITC are also available on the UN website.

For the tables that use the CPA nomenclature, the 2008 edition is used for the whole series.

Other elements influencing the comparability over time: The imports and exports of ships and oil platforms in the statistics were extended in 1991 to include transactions that concern vessels under a foreign flag with a Norwegian-registered holding company. Furthermore, the definition of developing countries was changed in 2007, which affects the figures slightly.

The data material in the External trade in goods statistics is very extensive, and at the most detailed level the quality can vary. The quality will be affected by errors in reported data. Collection errors can occur from incorrect values, incomplete information or misinterpretation of data on customs declarations. Examples of this are use of wrong commodity number, country code, county of production or incorrect quantity.

Users of the statistics must therefore consider this when using detailed figures. Commodity numbers with little trade, measured in value, will be more prone to errors than commodity numbers with a lot of trade. Furthermore, the statistical variables weight and quantity have poorer quality than value. Goods with restrictions and trade to particular countries are, however, monitored more closely.

Statistics Norway is required to report detailed figures to Eurostat, and therefore finds it appropriate that the figures are also published to Norwegian users, despite the varying quality.

Other sources of error include late updates, which sometimes occur for the trade with ships and oil platforms. The rules for customs clearance, which entail exemption from the duty to declare for goods with value below a set value, lead to lower figures than the real trade. This is not the best solution for preparation of the national accounts, where these lacking figures are estimated separately.

The calculation of seasonally-adjusted data beyond the usual seasonal effects is adjusted for the fact that trade is distributed unevenly across the week and taking into account the effect of Easter, as described in the section of estimations. However, beyond this nothing in particular is done. Therefore, it would be natural to assume that the seasonally-adjusted December figures should be interpreted more cautiously because of the Christmas holidays.

The export trade statistics is a full count, so there are therefore no non-response errors or sampling errors.

When publishing statistics for a new month of measurement, figures for the previous months in the same year are also revised. Furthermore, the figures for all months of the year are revised twice more: First time in May the following year (year t + 1), before final figures are published in May one year later (year t + 2). See External trade in goods tables in the Statbank in for more information about the size of the revisions.

At the aggregate level, the difference between preliminary monthly figures and revised figures published in May year t + 1 will normally be small for imports and mainland exports. At a detailed level and for individual months, the revisions can be of greater importance, relatively speaking. For oil and gas, larger audits can be experienced, in particular, related to the calculation of oil and gas prices. The revisions of the final publication in year t + 2 will normally be of minor importance, but in some cases relatively significant corrections can also be uncovered in this time perspective.

Monthly and quarterly time series are often characterised by considerable seasonal variations, which might complicate their interpretation. Such time series are therefore subjected to a process of seasonal adjustment in order to remove the effects of these seasonal fluctuations. Once data have been adjusted for seasonal effects by X-12-ARIMA or some other seasonal adjustment tool, a clearer picture of the time series emerges.

More information on seasonal adjustment, metadata on methods.

Many commodities in external trade in goods have a pattern that appear repeatedly at about the same time every year.

An example of seasonal variation for external trade in goods is the export of fish, where the seasonal pattern is influenced by fish species having different fishing seasons and international demands. The export of skrei is highest in February and March, and the salmon export highest towards the end of the year. Another example is the export of natural gas, where the amount exported varies with the winter and summer seasons, due to different demand from European countries.

To be able to follow the underlying development from month to month without being influenced by such variations, the figures are seasonally adjusted.

Pre-treatment is an adjustment for variations caused by calendar effects and outliers.

We follow the European Statistical System (ESS) guidelines on seasonal adjustment (ec.europa.eu) as far as possible. When background information for an outlier is available, the outlier is included as an explanatory variable (regressor) in the model.

For handling the corona crisis in the seasonal adjustment, we follow Eurostat's guidelines (available on Eurostat's web site as PDF), which state that the effect of the corona crisis should not be included in the basis for estimation of seasonal factors.

Calendar adjustment

Calendar adjustment contains both adjustments for trading days effects and for moving holiday effects. Adjustments for trading days means that we adjust the raw data so that both the number of working days since the composition of these can vary from period to period.

Calendar adjustment is carried out on all series that show a significant and plausible calendar effect with RegARIMA procedure (a regression model where the noise term is modelled by an ARIMA model).

Calendar regression variables are processed in accordance with the Norwegian holidays and public holidays.

Methods for trading/working day adjustment

RegARIMA correction is used – in this case, the effect of trading days is estimated in a RegArima framework. The effect of trading days can be estimated by using a correction for the length of the month or leap year, regressing the series on the number of working days etc. In this case, the residuals will have an ARIMA structure.

Correction for moving holidays

Test for moving holidays’ effect of Easter holiday after the X-13-ARIMAs impact modell. Very few series have a significant moving holidays’ effect.

National and EU/euro area calendars

Use of the Norwegian calendar, which considers Norwegian working days and public holidays.

Treatment of outliers

Outliers, or extreme values, are abnormal values of the series.

Series are inspected to identify outliers. Once identified, outliers are explained/modelled using all available information. Outliers for which a clear interpretation exists (changed methodology in data collections, special events and consequences of government policy changes etc.) are included as regressors in the model.

The identified outliers are predefined in pre-adjustment’s model. Program will only search for the last year for outlier. They are detected automatically by the seasonal adjustment tool. The outliers are removed before seasonal adjustment is carried out, and then reintroduced into the seasonally adjusted data.

Model selection

Pre-treatment requires choosing an ARIMA model, as well as deciding whether the data should be log-transformed or not.

Model selection is automatic, by JDemetra+’s established routines.

Decomposition routine

The decomposition routine specifies how the trend, seasonal and irregular components are decomposed. The most common decompositions are the multiplicative, additive or log additive.

We have used an automatic decomposition scheme selection in JDemetra+, and multiplicative decomposition is used for most series in this model.

Choice of seasonal adjustment approach: X-13 method in JDemetra+ (github.io)

Consistency between raw and seasonally adjusted data

In some series it is preferred that, for example, the sum of monthly seasonally adjusted figures for a year should be identical to the sum of monthly figures in the original raw series.

For the external trade statistics, no consistency conditions are applied.

Consistency between aggregate/definition of seasonally adjusted data

In some series, consistency between seasonally adjusted totals and the original series is imposed. For some series there is also a special relationship between the different series, e.g. GDP which equals production minus intermediate consumption.

Due to indirect approach for aggregates, the consistency is automatically fulfilled.

Direct versus indirect approach

Direct seasonal adjustment is performed if every time series, including aggregates, are seasonally adjusted on an individual basis. Indirect seasonal adjustment is performed if the seasonally adjusted estimate for a time series is derived by combining the estimates for two or more directly adjusted series.

A total of 38 seasonally adjusted series are published. Main aggregates such as total imports, total exports, mainland exports are seasonally adjusted indirectly. Furthermore, we have seasonally adjusted import and export series according to the one-digit SITC. All these series are value series. We find seasonal patterns for all these series, except SITC9 where series are adjusted mostly for calendar effects and the results from seasonal adjustment is uncertain.

The important export commodities crude oil, natural gas in gaseous form, fish (SITC03), metals other than iron and steel (SITC68) and electricity (SITC35) are seasonally adjusted for both value and volume, where volume series are direct adjusted and value series are indirectly adjusted.

Horizon for estimating the model and the correction factors

When performing seasonal adjustment of a time series, it is possible to choose the period to be used in estimating the model and the correction factors. Correction factors are the factors used in the pre-treatment and seasonal adjustment of the series.

The time series from January 2011 on is used to estimate the model and the correction factors.

General revision policy

Seasonally adjusted data may change due to a revision of the unadjusted (raw) data or the addition of new data. Such changes are called revisions, and there are several ways to deal with the problem of revisions when publishing the seasonally adjusted statistics.

In accordance with recommendations from the ESS guidelines, the models behind the seasonally adjusted figures will be subject to a thorough review once a year.

Concurrent versus current adjustment

Partial concurrent adjustment: the model is identified and estimated yearly, while filters, outliers (last year) and regression parameters are re-identified and estimated continuously as new or revised data become available.

Horizon for published revisions

The revision period for the seasonally adjusted results is limited to 4 years prior to the revision period of the unadjusted data, while older data are frozen.

Evaluation of seasonally adjustment data

A detailed set of graphical, descriptive, non-parametric and parametric criteria defined to assess the relevant characteristics of seasonally adjusted data are used.

Quality measures for seasonal adjustment

A set of available diagnostics within the seasonal adjusted tools and graphical capabilities are used.

A table containing selected quality indicators for the seasonal adjustments is available. The table is available here: indicators_seasonaladjustment (XLSX)

More information about the quality indicator is found here: metadata on methods: seasonal adjustment (PDF).

Seasonal adjustment of short time series

All series are sufficiently long to perform an optimal seasonal adjustment.

Treatment of problematic series

Following the Corona crisis that started in the 1st quarter of 2020, the seasonal adjustment follows the Eurostat guidelines for how to treat this extraordinary event. The result being that the effect of the Corona crisis is not a part of the foundation of the seasonal patterns. We assume that the seasonal pattern is unchanged, and that we correct for the systematic seasonal variation calculated on data before the corona crisis.

Data availability

Unadjusted figures (original series or raw data) and seasonally adjusted are available.

• Seasonal adjustment: general information (PDF)
• ESS guidelines on seasonal adjustment (Eurostat's web site)