|
For any question on data and metadata, please contact: Eurostat user support |
|
|||
1.1. Contact organisation | Statistics Sweden |
||
1.2. Contact organisation unit | Unit for Innovation, Business sector production and Research |
||
1.5. Contact mail address | Statistics Sweden Solna strandväg 86, 171 54 Solna |
|
|||
2.1. Metadata last certified | 15/05/2023 | ||
2.2. Metadata last posted | 15/06/2023 | ||
2.3. Metadata last update | 15/06/2023 |
|
|||
3.1. Data description | |||
Industrial Production Index, IPI. The main purpose of the industrial production index is to measure the change in the production of all the sectors of the entire industry. |
|||
3.2. Classification system | |||
NACE Rev. 2. |
|||
3.3. Coverage - sector | |||
The survey covers NACE Rev. 2. Sections B, C and D. The population of reference is based on kind-of-activity belonging to enterprises with 10 or more employees. |
|||
3.4. Statistical concepts and definitions | |||
The Industrial Production Index shows the changes in value added. The index is estimated using following data:
|
|||
3.5. Statistical unit | |||
Reporting unit: Kind-of-activity units. Observation unit: Kind-of-activity units. |
|||
3.6. Statistical population | |||
The target population for the surveyed enterprises is kind-of-activity units in NACE Rev. 2. sections B and C. The frame for identifying the population is the Swedish Statistical Business Register (SBR). The sample population is drawn in March. The target population consists of approximately 50 000 enterprises while the sample populations is approximately 2 100 Kind-of-Activity-Units. Demographic changes are constantly maintained in the sample population where a change has to be entered into the SBR in order to be entered in the sample population. Demographic changes are e.g. bankruptcy of an enterprise, the purchase or selling of an enterprise in sample population, mergers etc. |
|||
3.7. Reference area | |||
All regions of Sweden are covered. |
|||
3.8. Coverage - Time | |||
Data are available from 1968. However changes in methods, population definitions and base year have lead to breaks in the time series. Currently data are available from 2000 until present time in price level 2015 and in gross, calendar and seasonally adjusted data. |
|||
3.9. Base period | |||
Base year is 2015 |
|
|||
The final result is an index with base year equal to 100. Data used as input are deflated turnover in SEK and production volume in quantitites. |
|
|||
The reference period is calendar month. |
|
|||
6.1. Institutional Mandate - legal acts and other agreements | |||
Official statistics in Sweden are governed by the Official Statistics Act of 2001 (2001:99) and by the Statutes of Statistics Sweden (SCB-FS 1997:27). |
|||
6.2. Institutional Mandate - data sharing | |||
Data is transmitted monthly to the UN, IMF and EUROSTAT |
|
|||
7.1. Confidentiality - policy | |||
Statistics Sweden has policies in place in order to minimize the risk of disclosure and/or damage dealt to respondents in the survey. The main rule is that groups with less than three enterprises are always treated as confidential. However this criteria may be weak in certain situations. Therefore Statistics Sweden has developed a program that estimates the probability of disclosure in each of the reporting groups. Based on the estimated probability of disclosure the potential damage of such an exposure is assessed by the statisticians involved in the production of statistics. |
|||
7.2. Confidentiality - data treatment | |||
If data is under risk of disclosure, the first measure is cell suppression. If the risk of disclosure continues to recur in a certain cell, the second measure is aggregation of two or more similar cells. |
|
|||
8.1. Release calendar | |||
Dates for the monthly statistical releases are determined one year in advance. The dates can be found on Statistics Sweden's website, www.scb.se. |
|||
8.2. Release calendar access | |||
Anyone can access the release calendar on Statistics Sweden's website, Statistics Sweden's release calendar. |
|||
8.3. Release policy - user access | |||
Release is simultaneous to all interested parties. |
|
|||
The frequency of dissemination is monthly to all users. |
|
|||
10.1. Dissemination format - News release | |||
There is no news release |
|||
10.2. Dissemination format - Publications | |||
The data is not disseminated in any publication. |
|||
10.3. Dissemination format - online database | |||
Data is published in Statistics Swedens statistical databases on a pre-determined release calendar. The data base is available on-line to everyone. The statistical database is found here. The data is released in the database for the total industry, main industrial groupings and industry sub-sectors. Data is released in unadjusted figures, calendar adjusted figures, seasonally and calendar adjusted figures and trend figures. Data is released in monthly figures. |
|||
10.4. Dissemination format - microdata access | |||
Micro-data that makes it possible to identify individual objects is not publicly released. Statistics Sweden performs on request special processing of primary materials from previous surveys. Researchers may apply to Statistics Sweden to get access to de-identified micro data for own processing purposes. |
|||
10.5. Dissemination format - other | |||
Final data is sent to the National Accounts Department within Statistics Sweden to be used in the compilation of quarterly national accounts. Final data is also sent to Eurostat to be used in the compilation of European aggregates and to be released as national data. |
|||
10.6. Documentation on methodology | |||
Documentation about the survey is available in the form of Kvalitetsdeklaration (Quality Declaration) and SCBDOK (Documentation of statistics). Information about the final observation registries is stored in Statistics Sweden's database MetaPlus. All documentation is available in Swedish on Statistics Sweden's website. |
|||
10.7. Quality management - documentation | |||
The European Statistics Code of Practice compiles the criterias for the quality assessment of the official statistics of Sweden. |
|
|||
11.1. Quality assurance | |||
Statistics Sweden uses the EFQM Excellence Model from the European Foundation for Quality Management as a framwork for quality assurance. Statistics Sweden is certified under the ISO 20252 standard. |
|||
11.2. Quality management - assessment | |||
Overall Statistics Sweden has a high level of quality. The Industrial Production Index is highly relevant both as an input to quarterly national accounts and stand-alone. The level of accuracy is high. Comparability over time is somewhat affected by methodological changes to increase the accuracy. It is comparable with other statistical domains, primarily the quarterly national accounts. |
|
|||
12.1. Relevance - User Needs | |||
Primary users and their needs are:
Secondary users and their needs are:
|
|||
12.2. Relevance - User Satisfaction | |||
Statistics Sweden has general satisfaction surveys but no specific survey for the short-term statistics. Views and opinions of the users are collected through user councils with 2 to 4 meetings per year. The members of the councils represent the users of the statistics and the council is funded by grants produced by Statistics Sweden. The council has an advisory role regarding questions concerning new statistics, development and improvement of existing statistics and priorities for the coming financial year. Direct contact with users is also a source for evaluating user satisfaction. |
|||
12.3. Completeness | |||
Statistics Sweden has started to collect for NACE 09 - Mining support service activities. However quality needs to be evaluated before statistics can be disseminated. Eurostat's requirements in terms of time series are fulfilled. |
|
|||
13.1. Accuracy - overall | |||
There are a number of sources for inaccuracy in the Industrial Production. This is an attempt to structure the possible sources for inaccuracy and to qualitatively describe the effects. Coverage: The turnover survey covers the whole industrial sector i.e. kind-of-activity units involved in industrial activities. The frame is constructed using a "frozen" version of the business register, so enterprises entering into the business register after this freezing moment are considered undercoverage, and enterprises exiting the frame after this moment are considered overcoverage. Sampling: The sample method is a stratified random sampling. Stratification variables are NACE Rev.2 activity and size of yearly turnover. The sample consists of seven size classes where size classes 1-4 are sampled, and 5-6 and 8 are censused. Size classes 1-4 consist of the smallest units and 5-6 of the largest units. Size class 8 consists of complex units consisting of two or more legal units. Size classes are created using the Dalenius-Hodges rule. Sample size in each stratum is determined using Neyman-allocation. Nonresponse: Nonresponse is handled by imputation methods for both item and object nonresponse. The imputation method consists of a list of ranked alternatives where the first alternative is used if the input value meets a set of criterias. If the first alternative is not possible due to failure to meet the criterias the next alternative is used and so on. The process is automated using the BANFF-system in SAS. Measurement: As always there may be measurement errors with sampling. The statistics is cross-checked with multiple sources such as VAT-data, trade statistics, SBS, etc. in order to identify measurement errors and correct them. Questionnaire data for units in size classes 1-4 are replaced with VAT-data on a quarterly basis for all units in the frame which decreases measurement errors. However the VAT-data also contain measurement errors and may have periodicity problems. Since each unit only represents itself the measurement errors are limited when comparing to a "normal" estimation. Data processing: Turnover is deflated using producer price indices and weighted using value added from the quarterly national accounts. Errors and inconsistencies in those data sources may affect accuracy of the Industrial Production Index. Model assumptions: The frame is constructed using a cutoff where units over the cutoff that together have the top 95 % of the yearly turnover. The turnover for the units under the cutoff are assumed to have the same devlopment in turnvover as the units over the cutoff. |
|||
13.2. Sampling error | |||
The sample method is a stratified random sampling. Stratification variables are NACE Rev.2 activity and size of yearly turnover. The sample consists of seven size classes where size classes 1-4 are sampled, and 5-6 and 8 are censused. Size classes 1-4 consist of the smallest units and 5-6 of the largest units. Size class 8 consists of complex units consisting of two or more legal units. Size classes are created using the Dalenius-Hodges rule. Sample size in each stratum is determined using Neyman-allocation. The turnover survey is a probability sampling survey for monthly estimates but a census on a quarterly basis. On a quarterly basis, data for units that is sampled using probability sampling is replaced by VAT-data for the part of the frame in size classes 1-4. Data for censused size classes are never replaced with VAT-data. Monthly estimates based on survey data only is then replaced. The sampling error for the "census" results on a quarterly basis is zero since it is a census. The sampling error for the estimates based on survey data can be quantified. |
|||
13.3. Non-sampling error | |||
The turnover survey covers the whole industrial sector i.e. kind-of-activity units involved in industrial activities. But the frame is constructed using a "frozen" version of the business register, so enterprises entering into the business register after this freezing moment is considered undercoverage, and enterprises exiting the frame after this moment is considered overcoverage. The survey instrument is an electronic questionnaire. The questionnaire has been evaluated using cognitive interviews and expert evaluations in order to reduce the possibilities for non-sampling errors. Nonresponse is handled by imputation methods for both item and object nonresponse. The imputation method consists of a list of ranked alternatives where the first alternative is used if it meets a set of criterias. If the first alternative is not possible due to failure to meet the criterias the next alternative is used and so on. The process is automated using the BANFF-system in SAS. Turnover is deflated using producer price indices and weighted using value added from the quarterly national accounts. Errors and inconsistencies in those data sources may affect accuracy of the Industrial Production Index. The frame is constructed using a cutoff where units over the cutoff that together have the top 95 % of the yearly turnover. The turnover for the units under the cutoff are assumed to have the same development in turnvover as the units over the cutoff. All nonsresponses are handled by imputation so the imputation methods affect the accuracy of the final estimates. Production volumes and hours worked are always in place at time of the results estimation. Mean weighted response rate for total industry at first estimate of a reference period in 2022 was 84 percent. Turnover is used as weights. |
|
|||
14.1. Timeliness | |||
Data is released no later than one month plus 10 days after the end of the reference period. All aggregation levels and sub-sectors are released at the same time. Data collection Turnover: Questionnaires are sent to the reporting units at the end of the month of the reference period asking for replies by the 15th of the following month. Data collection is normally closed 3 days before publishing. Production volumes: Questionnaires are sent to the reporting units at the end of the month. The reporting unit has approximately 4 weeks to reply. |
|||
14.2. Punctuality | |||
All releases have been done on time. |
|
|||
15.1. Comparability - geographical | |||
The same statistical concepts are applied in the entire Swedish national territory. No geographical discrepancies exist. |
|||
15.2. Comparability - over time | |||
Comparability over time is mostly affected by changes in NACE classification system. The transition from NACE Rev.1.1 to NACE Rev.2 meant that other publishing industry than recycling industry is not included in the industry. Another factor that affects the comparability over time is the transition from a fixed-base index to a chain index in 2005. In April 2015 Statistics Sweden changed the methodology mostly with respect to sampling and estimation. In order to avoid breaks, the time series were spliced together based on the information from collecting and estimating statistics with both old and new method. From start of the survey in 1968 a gradual transition has occured from the use of production volumes and hours worked to turnover as an input in the estimation. At the transition from NACE Rev.1.1 to NACE Rev.2 the Industrial Production Index was re-estimated back to 2000. In NACE Rev.1.1 time series exist from 1990M01 to 2008M12. In NACE Rev.2 time series exist from 2000M01 to current. In the time series no differences in length exist between monthly, quarterly and yearly data. |
|||
15.3. Coherence - cross domain | |||
The Industrial Production Index is comparable to the monthly surveys Index of Service Production and Orders and Turnover in Industry, as well as the quarterly surveys Industrial Inventories and Industrial Capacity Utilization due to similarities in terms of survey frame, sample drawing, estimation and model assumptions. The Industrial Production Index is used as an input in the quarterly National Accounts. The annual National Accounts use the annual Structural Business Statistics as an input. The coherence between IPI and the quarterly National Accounts is therefore high. Price statistics, PPI for domestic and export markets, are used in the Industrial Production Index to deflate turnover. Industrial turnover from the survey Orders and turnover in industry are used as an input in the Industrial Production Index. |
|||
15.4. Coherence - internal | |||
Input data for the Industrial Production Index comes from three different sources, production volume, hours worked and deflated turnover. The first step in index calculations differs depending on the type of data source. The following steps are carried out in the same way regardless of the data source. Seasonal adjustment may lead to some incoherence between indices on sub-sector level and aggregates on e.g. MIG-level. Seasonal adjustment is done after the index calculations and on an individual level which may lead to some incoherence between aggregates and the individual components, i.e. sub-sectors, making up an aggregate. |
|
|||
In total, Statistics Sweden estimates that 1000 working hours per year is associated with the work at Statistics Sweden for the collection, estimation, calculation and presentation of the Industrial Production Index. The main part of the index is calculated using data from industrial turnover collected by the survey on Orders and turnover in industry. The burden on respondents corresponding to the turnover part of that survey is estimated to approximately 5600 hours per year. The costs are associated both with national requirements and specific STS-requirements although the proportions are not known. |
|
|||
17.1. Data revision - policy | |||
Routine revisions are made for two to four months prior to the current month depending on which month in the quarter the reference month is. The whole previous quarter is revised in connection with the publishing of the first and second month in a quarter. Revisions for previous periods are released at the same time as the release of a new period. The same revision policy is applied nationally and in transmissions to Eurostat. Industrial Production Index is a chain-index, so no revisions occur due to regular base year changes (i.e. the rebasing that is carried out every fifth year due to STS requirements). Major revisions may occur due to methodological changes. Methodological changes are announced in advance. Non-scheduled revisions, i.e. unexpected revisions may occur due to errors discovered in the input data after results are considered definitive, if deemed necessary due to the magnitude of the error. Non-schedule revisions may also occur between normal releases if errors of a greater magnitude are discovered shortly after release. Vintages are stored but not published, they are however provided upon request. |
|||
17.2. Data revision - practice | |||
Data in the questionnaire is collected for the last three periods which introduces revisions. Enterprises revising previously submitted data also introduce revisions. Users are informed in advance of methodological and other changes which may lead to revisions. The information is included in the press release the month before the changes will come in to effect. Users are not informed in advance of the actual revisions. No special revision calendar exists and revisions are published at the same time as figures for new months. The same revision policy is applied to data released nationally and transmitted to Eurostat. Using the Eurostat guidlines on Quality Indicators MAR (Mean Absolute Revision) and MR (Mean Revision): The computed values for the last 36 monthly (Jan 2020 - Dec 2022) for total industry (B-D) are: Growth rates for calendar/working day adjusted data series (YoY): MAR = 0.71 , MR = -0.37 |
|
|||
18.1. Source data | |||
Input data comes from two different sources. The sources and their corresponding part of the Industrial Production Index are: Enterprise survey on turnover (95 procent): Stratified random sample survey on kind-of-activity units. Approximately 98 percent of the enterprises use an electonic questionnaire and approximately 2 percent use paper questionnaires. Stratification variables are NACE Rev.2 activity and size of yearly turnover. Size classes are created using the Dalenius-Hodges rule. Sample size in each stratum is determined using Neyman-allocation. The two largest strata are censused and the four smallest one are sampled. Complex units i.d. Enterprises consisting of two or more legal units are always censused. The frame and sample population is updated in April each year. Production volumes: Collected through various trade associations and reported to Statistics Sweden. |
|||
18.2. Frequency of data collection | |||
All data is collected on a monthly basis. All data refers to the full reference period. |
|||
18.3. Data collection | |||
Enterprise survey on turnover: Collected through random sampling survey. Most enterprises (approximately 98 percent) use an on-line questionnaire although a few (approximately 2 percent) use a paper questionnaire. New enterprises in the survey are contacted through regular mail with information and login information. Each month, after the end of the reference period, login information is sent to the enterprises through regular mail reminding them of the survey. A reminder is sent out around 15 days after the first contact. Import and/or larger enterprises are reminded especially if they have not yet responded at the end of the collection phase. During the production phase enterprises are contacted in order to get data from important and/or larger enterprises. Contact is primarily through phone and e-mail. The survey has a sanction system with a penalty for not responding. |
|||
18.4. Data validation | |||
Data is "checked-in" in multiple steps:
Final results are "checked-out" in multiple steps:
|
|||
18.5. Data compilation | |||
Missing observations from unit and item non-responses are delt with by using automatic imputations carried out by the estimation programs. The estimator is an Horvitz-Thompson type of estimator combined with VAT-information. Sampled enterprises are replaced with turnover based on VAT-data when quarterly turnover data is available. Censused units are never replaced with turnover from VAT. Turnover based on VAT does not only apply sample units in stratas that are sampled, but all units in the frame so the survey is a census on a quarterly basis. The index method is a chained Laspeyre-index with weights updated each year using annual overlap. Weights used in the index calculations are value added from the Structural Business Statistics modified by the National Accounts. Weights are updated yearly. A program developed by Statistics Sweden fetches data from the internal databases and converts it to sdmx-fil in order to transmit the data to Eurostat. |
|||
18.6. Adjustment | |||
Monthly turnover are deflated using monthly Producer Price Indices on a three digit level for both domestic and export markets. Production volumes are weighted using the annual PPI for the previous year before the index is calculated. The results are not adjusted by a factor derived from (forecasted) results from an annual benchmark survey. Prices are reported to Statistics Sweden in SEK. Any conversion from foreign currency is carried out by the respondents. No sub-indices are compiled. Calendar adjustment A country specific calendar i used. Moving holidays that are adjusted are Easter and the feast of the Ascension. Leap year effects are adjusted for. Calendar effects are adjusted for regARIMA. Other pre-adjustent RegARIMA to detect all types of outliers (AO, TC, and LS). LS outlier in Dec 2008 because of the finance crisis. Seasonal ajustment The software used is X-12-ARIMA in SAS® system (PROC X12) with SAS 9.3. The model/filter selection is manual for ARIMA models and automatically chosen (default) filter. Models and parameters are reviewed once a year. However partial concurrent adjustment is applied so parameters are re-estimated every time a new round is compiled. The seasonal adjustment is applied to the whole time series back to January 2000. Minor revisions may therefore occur over the whole series. The seasonal adjustment composition for Industrial Production Index is a multiplicative model. Otherwise both additive and multiplicative models are used for different series. The models are checked for adequacy. The critical value for outlier detection is predefined (3.5). The filter length is automatically chosen. No seasonal breaks are entered into the series to methodological changes. Series are directly adjusted. Since series are directly adjusted some inconsistencies may exist where the development in an aggregate may not be the exact sum of the underlying indices. |
|
|||
No further comments. |
|
|||
|
|||