Reference metadata describe statistical concepts and methodologies used for the collection and generation of data. They provide information on data quality and, since they are strongly content-oriented, assist users in interpreting the data. Reference metadata, unlike structural metadata, can be decoupled from the data.
Industrial, Construction and Energy Statistics Department (51) Industrial Statistics Unit (5101)
1.3. Contact name
Confidential because of GDPR
1.4. Contact person function
Confidential because of GDPR
1.5. Contact mail address
Český statistický úřad Na padesátém 81 100 82 Praha 10
1.6. Contact email address
Confidential because of GDPR
1.7. Contact phone number
Confidential because of GDPR
1.8. Contact fax number
Confidential because of GDPR
2.1. Metadata last certified
7 May 2025
2.2. Metadata last posted
7 May 2025
2.3. Metadata last update
7 May 2025
3.1. Data description
Industrial Production Index (IPI)
The indicator is primarily based on specific monthly business survey.
3.2. Classification system
NACE Rev. 2.
3.3. Coverage - sector
NACE Rev. 2 Sections B to D (except Group 35.3). The results cover the whole population of enterprises, what represents institutional sectors of non-financial corporations and household firms.
3.4. Statistical concepts and definitions
The IPI is an business cycle indicator which measures monthly changes in the price-adjusted output of industry.
The index is compiled in accordance with concept and definitions from STS Methodological Manual (Methodology of short-term business statistics: Interpretation and guidelines). In order to approximate the net output the following inputs are used: Production quantities by product (NACE Divisions 05, 06, 12, 19 and 35) and turnover from industrial activity at constant prices.
3.5. Statistical unit
Reporting and observation unit is enterprise as identified by national business register. In order to approximate the KAU the data are adjusted for non-industrial activities, important KAUs are surveyed separately.
3.6. Statistical population
Enterprises classified in NACE B-D (except Group 35.3) regardless the size class. The population is defined upon the national business register and is continuously updated (incl. demographic changes). An overview on the structure of the population below represents the number of active units in 2024.
NACE Rev.2 Sections
Number of active units in 2024
B
495
C
182 908
D
12 603
Total
196 006
3.7. Reference area
Czech Republic.
3.8. Coverage - Time
Length of time series: Year 2000 - Presence: - The infomation is valid for all aggregates required by EBS-Regulation. - The data available nationally are identical to those trasmitted to Eurostat.
3.9. Base period
Base year = reference year 2021: - There is no difference between national presentation and the data trasmitted to Eurostat.
Index (%).
Calendar month.
6.1. Institutional Mandate - legal acts and other agreements
Act No. 89/1995 Coll. on the state statistical service of June 15, 1995, as amended (does not refer to any specific indicator).
Regulation (EU) 2019/2152 of the European Parliament and of the Council of 27 November 2019 on European business statistics.
Commission Implementing Regulation 2020/1197 laying down technical specifications and arrangements pursuant to Regulation (EU) 2019/2152 of the European Parliament and of the Council on European business statistics repealing 10 legal acts in the field of business statistic (General Implementing Act).
Data sharing with international organizations: EUROSTAT, IMF, UN, OECD (tables predefined by the organization).
7.1. Confidentiality - policy
The confidentiality policy is based on: 1. Act No. 89/1995 Coll. on the state statistical service as amended (assures protection of individual data and implies the active approach to data protection), 2. Law on protection of personal data, 3. The internal Directive No. 3/2015 of the Czech Statistical Office (the defines the detailed rules, which are seen as confidential).
7.2. Confidentiality - data treatment
Data are only published down to an activity level at which they are not confidential, i.e. confidential cells are suppressed from publication. The data allowing direct or indirect identification of individual enterprise can be published under written permission of all related subjects.
8.1. Release calendar
For each year the Catalogue of Products containing Calendar of News Releases is published at the end of the preceding year. The data are part of "Industry" News Release.
8.2. Release calendar access
Release calendar is available at the CZSO website.
The data are released simultaneously to all interested parties by issuing the News Release "Industry" at 9:00 AM on the scheduled day of release. No user has pre-release access to the data. Transmission of the data to Eurostat begins on the day of national publication.
Monthly. - There are no differences between transmissions to Eurostat and national dissemination.
10.1. Dissemination format - News release
The data represents the key component of regular monthly "Industry" News Release, which is published on the CZSO Website. The News Release is amended by set of tables, charts and by updated time series starting from 2000 (see point 10.5).
According to the national legislation the access to the microdata is allowed for scientific purposes only. Each request is decided on individual basis by the relevant expert group.
10.5. Dissemination format - other
Other formats: - The most comprehensive infomation is available in the Public Database. It contains tables with selection of parameters and links to selected time series for download in xls format. - The data are regularly transmitted to Eurostat.
Important aspects and procedures of quality management within statistical production process as well as methodology are documented by the internal Statistical Metadata System (also in the form of so-called Technical projects), event. instructions for surveyors that are available for every statistics produced.
Technical projects in general contain description of the primary input data, data sources, definition of population, sampling procedure, processing technology, requirements for data transfer, control relations and checks, automatic corrections, derived indicators, definition of organization responsibilities, description of controls and relations, methodology of imputations, definition of outputs, description of input, source and user databases, backups, schedules of project and program preparation, schedules of preliminary and final processing, etc. In the CZSO there is a process of technical project preparation that is coordinated by the Department for Methodology of Performing Statistical Processing. For these purposes every year the so-called Schedule of technical preparation of statistical surveys is discussed in the advisory bodies of the President, where activities, terms and responsible units are specified.
As regards documentation accessible to the wide public, an elementary description of the methodology is published in the methodological notes on the CZSO website (topic: Industry - see the link below). Advanced information on methodology appears occasionally in specialised periodicals, detailed methodology of the IPI calculation is available in Czech language only (see the link below).
The whole process of data production is driven by internal documentation (Technical project and the Statistical Metadata System) to which all interested parties has an access (see the point 11.1).
Audit reports including all the test and analysis carried out within the audit are stored and the public parts are available on request (see also quality management).
11.1. Quality assurance
The key components of the quality assurance are the following: 1. Strategic documents of the CZSO: Mission, Vision and Priorities of the CZSO, Quality commitment, Priority tasks defined for each year. 2. Irregular methodological audits represents an important tool in quality control. Audits are usually carried out by expert group comprising of internal and external experts (NA experts, representatives of universities, ministries etc.). The IPI was audited in 2013-2014. 3. Important aspects and procedures of quality management within statistical production process are documented in by internal Statistical Metadata System, including so-called Technical projects, event. by instructions for surveyors that are available for every statistics produced.
Assessment of the quality uses the following components: 1. Standard STS compliance assessments carried out by Eurostat. This assessment is used internally as well. 2. The methodological audit (see also Quality assurance) includes inter alia different quality analysis, e.g. benchmark analysys, response analysis, in case of leading indicators, their ability to predict the changes in the economy. 3. Regular response rate measurement. 4. Regular confrontation of estimates with later available administrative data or comparable data acquired or estimated from statistics with different periodicity. 5. Regular measurement of other qualitative indicators (e.g. coefficient of variation).
12.1. Relevance - User Needs
The following users can be identified: 1. External users: 1.1 Eurostat, 1.2 Ministry of Industry and Trade, Czech National Bank, other governmental institutions, other international organizations (IMF, UN, OECD), 1.3 Employers' associations, trade unions, media, students, wide public, 2. Internal users: Quarterly National Accounts Department.
The CZSO primarily fulfill its commitments towards the European Commission, which represents the high priority user. However, depending on the capacity available the CZSO treats all the users equally.
12.2. Relevance - User Satisfaction
No users satisfaction surveys are carried out exclusively for STS indicators on the national level. However, CZSO is constantly in contact with different groups of users, e.g. Ministry of Industry and Trade, Czech National Bank, different associations, internal users etc. Formal meeting with users is organized regularly by the General Methodology Department.
12.3. Completeness
The CZSO provides Eurostat with all data requred by the EBS Regulation.
13.1. Accuracy - overall
Overall assessment: - With regards to the reasonable response rate the first estimates can be seen as not biased. - The main problem is the lack of information related to the smallest units in the population, what results also in the high misclassification. However, in case of Industry the problem is less significant than in other branches.
13.2. Sampling error
Sampling error indicators are foreseen to be a standard component of the SMS Quality system, which is currently being developed.
13.3. Non-sampling error
The main problems related to non-sampling errors are the the following: 1. Administrative data available for the smallest units in the frame (i.e. natural persons and legal persons without employees, which are not VAT payers) are outdated at the moment when the sample is drawn (data refers to R-1 rather than R). No other data source for activity assessment is available. 2. Relatively high degree of misclassification particularly of small units (natural persons and legal persons without employees, which are not VAT payers). The details are available in the SBS Quality report. 3. Non-response. However, it is not regarded as serious problem. See below for details. 4. Due to tough deadlines the respondents do not always have their book-keeping regarding the reported month closed when submitting data. It often leads to revisions in the next periods. Also mistakes in book-keeping can cause later revisions. 5. Organisational changes (mergers, splits, births and deaths) in reporting units and in the population as a whole during the year. The BR is relatively slow, what increases an importance of feedback from infraannual surveys. 6. General problem: According to the CZSO instructions the required financial indicators should be reported according to the Czech Accounting Standards (CAS). However, especially multinational, global enterprises keep accounts according to international accounting standards (IFRS, GAAP etc.). They are not always able to provide CZSO with the data in conformity with the CAS in time. It often leads to later revisions especially in case of quarterly surveys where more detailed infomation is collected (limited impact on turnover and similar indicators collected on monthly basis).
The non-response rate is approx. 18% at the time when the survey is closed for a reference month (i.e. preliminary data published 37th day after the end of the reference period). The non-response rate of final data reaches 5%.
14.1. Timeliness
The target deadline for the release is 37th day after the end of the reference period. Exact dates (see the Release calendar) take into the account the distribution of working days in the particular period. These deadlines allow to meet requirements of both EU legislation and other users' requirements.
14.2. Punctuality
The data are always published at the 9:00 AM of the day announced by the Release Calendar. The CZSO does not register any deviation from this practice.
15.1. Comparability - geographical
Not applicable at MS level. The same concepts are used within the whole national territory.
15.2. Comparability - over time
Relevant time series are available starting from 2000. - There is no difference in comparability or acessibility of the data for various frequencies.
15.3. Coherence - cross domain
The module in shared in all STS monthly surveys and as such the coherence of the results is assured. Checks between annual and monthly/quarterly data are performed regularly as part of validation of the survey with higher frequency of data collection (e.g. comparison of monthly STS survey on Industry with PRODCOM data).
Comparison within IPI and quarterly data for QNA is standard part of data processing.
Assessment of coherence and comparability with National Accounts is standard part of methodology audits. The results from the latest methodological audit (2013/2014) shows good correlation at the level of industry total.
15.4. Coherence - internal
There are no problems regarding internal consistence of the time series in question. In case of seasonal and calendar adjustment the direct method is used. It affects coherency between different aggregation levels.
The CZSO performs the burder measurement by statistical survey (questionnaire). With regard to the indirect relation between input and output data the precise distribution of the overall burden by the individual indicators is not possible. The latest information available is given below in annexes. The data refers to the whole related survey.
Data revisions follow the Revision Policy of the CZSO, which represent one of the key documents of the office. It is accessible via the CZSO website in both Czech and English. Calendar of standard revisions is being prepared together with the release calendar.
The same revision policy is applied to STS data released nationally and transmitted to Eurostat.
Vintage database currently exists in the central processing database. The public database contains actual version of the published data. Outdated data versions are internally saved.
Monthly Revisions: When data for February of the year Y are released data for January of the year Y are revised. Main reasons here are specific issues related to the start-up of tasks for the new reference period (partial exchange of units of the sample, increased non-response, updates of registers (basis for the sampling) – the vast changes carried out during the year).
Quarterly Revisions: As the need may be revisions of monthly and quarterly data for the last elapsed quarter are carried out. Simultaneously, revisions of previous periods of the reference year may also be performed. All these ordinary revisions are released once a quarter elapsed along with the appropriate News Release for the first month of the current quarter. The revisions are based on knowledge and more accurate data from quarterly tasks.
Annual Revisions: 1st annual revision of monthly estimates for all reference periods of the last elapsed year (Y) is released along with the appropriate News Release for January. In order to make these estimates data from annual tasks for the reference year Y-1y are employed. As the need may 2nd annual revision of monthly estimates for all reference periods of the last elapsed year (Y-2y) is released. If there is no reason to do so, the revision is not carried out. If there are some, data from annual tasks for the same reference year (Y-2y) are used. Data released in the second year since the reference period has elapsed at the latest are considered to be finalised data.
Monthly unadjusted data: MR = 0,08, MAR = 0,10 Monthly adjusted data: MR = 0,08, MAR = 0,11
18.1. Source data
Type of data source: Statistical survey (Prum 1-12)
Frame on which the source is based: Business Register (RES); the population represents enterprises with main activitity in NACE Rev.2 Sections B to D.
Sample or census: Production quantities are surveyed exhaustively above the cut-off treshold (annex to the questionnaire PRUM 1-12). A stratified sample is used for the collection of the data on turnover (questionnaire PRUM 1-12).
Criteria for stratification: Characteristics used for stratification are the activity, number of employees and the turnover as registered by the VAT declaration.
Threshold values and percentages: All units belonging to NACE Divisions 05 to 35 with annual turnover greater than CZK 200 million are included in the sample (regardless of their size category), as are all units with 100 or more employees. The whole sample contains approximately 5 085 units. The percentages of the randomly choosen units with less than 100 employees is shown in the following table in annexes - note that the strata for 10*, 13* and 25* are for units for which classification is only available at the 2-digit level.
Frequency of updating the sample: Sample design can be updated once a year. Units included in the sample are updated monthly.
Questionnaires used in the survey: — Prum 1-12 (monthly collection of turnover). — Annex to Prum 1-12 (monthly collection of production quantities).
Data collection media: Electronic questionnaire, web questiononnaire, printed questionnaire. Starting from 2013 printed questionnaires are available only to those respondents which are impossible to contact electronically or upon request.
Non-response treatment: Phone, email or postal reminders. The activity starts the next day after the deadline of the questionnaire. Fines can be given to non-responding units. The related processes are driven by internal Directive.
18.4. Data validation
The data validation is driven by survey documentation (Technical project, SMS Ulohy), which decribes approx. 80 validation checks attributed to the processing of the questionnaire Prum 1-12. This contains: - Validity of each attribute collected within the questionnaire (e.g. enterprise's ID, NACE, etc.); - Logical checks between indicators inside the questionnaire; - Comparison over time (previsous period / the same period of the previous year); - Comparison between enterprises belonging to the same cathegory (NACE + size class, comparison of unit prices where applicable).
The validation between surveys is attributed to the surveys of higher frequency (quarterly, annual).
Plausibility checks (e.g. comparison with NA) are performed at ad-hoc basis. They are also an integral part of methodological audits.
Feedback from the validation on micro level is available also to reporting units in case of electronic data collection (electronic questionnaire, online data collection).
For the transmission to Eurostat the internal application SDMX-STS had been developed. Except encoding to Gesmes and communication with eDamis it assures also consistency with predefined structure of the data as well as appropriatate coding. As the application is part of the data production system, specific validation of the STS files is not needed.
18.5. Data compilation
Estimates for non-response: Not responding units are imputed. Estimation for non-response is based on the results of similar enterprises from the same strata. If the unit responded in previous period(s) then the estimations takes these results into account.
Estimates for grossing-up to population levels: The results are grossed up to the whole population.
Type of index: Laspeyres.
Method of weighting and chaining: Fixed base index - the weighting schemes have been used that were derived from the results of structural business statistics for the year 2021. When production quantities are used (see the point 3.4), the first stage weights up to 4 digits level are derived from production values in PRODCOM.
Change of the base year to 2021: Series were recalculated back to 2014 using the most up-to-date weighting scheme. Historical data were linked via annual overlap. These principles were not applied previously with regard to the NACE Rev.2 introduction.
Transmission of the data to Eurostat: The files are converted to SDMX using internally developed application
18.6. Adjustment
Deflation: For the relevant deflated turnover is used (see point 3.4). In such case deflation is performed separately for non-domestic and domestic component of the turnover using the producers prices and export prices. The deflation is carried out at the level of NACE divisions or classes (the level is chosed according to affiliation of the level to MIG).
Seasonal and working day adjusment is performed on the following series using JDemetra+ sofware (TRAMO-SEATS method): Totals - BTD, B_C, B_CXMIG_NRG NACE Sections - B, C, D NACE Divisions - B05, B06, B07, B08, B09, C10, C11, C12, C13, C14, C15, C16, C17, C18, C19, C20, C21, C22, C23, C24, C25, C26, C27, C28, C29, C30, C31, C32, C33 MIGs and their aggregates - MIG_CAG, MIG_COG, MIG_COGXFOOD, MIG_DCOG, MIG_ING, MIG_ING_CAG, MIG_NDCOG, MIG_NRGXE
For details on seasonal and working day adjusment please see the annex, which refers to the situation as of 2024.
The indicator is primarily based on specific monthly business survey.
7 May 2025
The IPI is an business cycle indicator which measures monthly changes in the price-adjusted output of industry.
The index is compiled in accordance with concept and definitions from STS Methodological Manual (Methodology of short-term business statistics: Interpretation and guidelines). In order to approximate the net output the following inputs are used: Production quantities by product (NACE Divisions 05, 06, 12, 19 and 35) and turnover from industrial activity at constant prices.
Reporting and observation unit is enterprise as identified by national business register. In order to approximate the KAU the data are adjusted for non-industrial activities, important KAUs are surveyed separately.
Enterprises classified in NACE B-D (except Group 35.3) regardless the size class. The population is defined upon the national business register and is continuously updated (incl. demographic changes). An overview on the structure of the population below represents the number of active units in 2024.
NACE Rev.2 Sections
Number of active units in 2024
B
495
C
182 908
D
12 603
Total
196 006
Czech Republic.
Calendar month.
Overall assessment: - With regards to the reasonable response rate the first estimates can be seen as not biased. - The main problem is the lack of information related to the smallest units in the population, what results also in the high misclassification. However, in case of Industry the problem is less significant than in other branches.
Index (%).
Estimates for non-response: Not responding units are imputed. Estimation for non-response is based on the results of similar enterprises from the same strata. If the unit responded in previous period(s) then the estimations takes these results into account.
Estimates for grossing-up to population levels: The results are grossed up to the whole population.
Type of index: Laspeyres.
Method of weighting and chaining: Fixed base index - the weighting schemes have been used that were derived from the results of structural business statistics for the year 2021. When production quantities are used (see the point 3.4), the first stage weights up to 4 digits level are derived from production values in PRODCOM.
Change of the base year to 2021: Series were recalculated back to 2014 using the most up-to-date weighting scheme. Historical data were linked via annual overlap. These principles were not applied previously with regard to the NACE Rev.2 introduction.
Transmission of the data to Eurostat: The files are converted to SDMX using internally developed application
Type of data source: Statistical survey (Prum 1-12)
Frame on which the source is based: Business Register (RES); the population represents enterprises with main activitity in NACE Rev.2 Sections B to D.
Sample or census: Production quantities are surveyed exhaustively above the cut-off treshold (annex to the questionnaire PRUM 1-12). A stratified sample is used for the collection of the data on turnover (questionnaire PRUM 1-12).
Criteria for stratification: Characteristics used for stratification are the activity, number of employees and the turnover as registered by the VAT declaration.
Threshold values and percentages: All units belonging to NACE Divisions 05 to 35 with annual turnover greater than CZK 200 million are included in the sample (regardless of their size category), as are all units with 100 or more employees. The whole sample contains approximately 5 085 units. The percentages of the randomly choosen units with less than 100 employees is shown in the following table in annexes - note that the strata for 10*, 13* and 25* are for units for which classification is only available at the 2-digit level.
Frequency of updating the sample: Sample design can be updated once a year. Units included in the sample are updated monthly.
Monthly. - There are no differences between transmissions to Eurostat and national dissemination.
The target deadline for the release is 37th day after the end of the reference period. Exact dates (see the Release calendar) take into the account the distribution of working days in the particular period. These deadlines allow to meet requirements of both EU legislation and other users' requirements.
Not applicable at MS level. The same concepts are used within the whole national territory.
Relevant time series are available starting from 2000. - There is no difference in comparability or acessibility of the data for various frequencies.
