Statistical processing

Source data

The frame population for these statistics are all kind of activity units (KAU's) with main activity in mining and quarrying or in manufacturing and having at least 10 employees or a yearly turnover over 100 mio. dkk. The population is selected annually, based on the employment registered for 3rd quarter of the year before the reference year. The information on industrial classification is based on the Statistical Business Register as of the beginning of the reference year. Data is collected from approx. 3,000 units who are part of the survey. For Manufacturers' sales of goods, revenue is reported across a number of categories: Sales of goods (on CN commodity codes), including production on own account, production outsourced to others and production as subcontractor, as well as industrial services. In addition, there are sales of trading goods and other turnover.

Frequency of data collection

The statistics are an annual survey, first collected in this form in 2025.

Data collection

Data are collected via an electronic reporting solution or through an upload solution using a standardized Excel spreadsheet. Guidance, etc., can be found on the statistics reporting page.

Data validation

The electronic reporting solution includes a number of automatic checks. This means that if a respondent is about to submit a report that differs significantly from what is expected (compared with their most recent report), they will immediately be asked to consider whether it is correct. Once the reports are received by Statistics Denmark, they are subjected to a number of checks, for example in relation to large fluctuations compared with the previous period or in relation to the typical price (value/quantity) for the commodity codes used. It is also checked whether the reported commodity codes are plausible in relation to the enterprise’s industry. The reported data are also compared with other comparable statistics, primarily ‘Production and turnover in manufacturing industries’ and ‘Purchases and sales by enterprises’. The latter is particularly useful for identifying cases of underreporting. Comparisons with other statistics are carried out both at the enterprise level and at the industry level.

Data compilation

After data validation, imputation is carried out for missing reports. Imputation means that missing data are replaced with plausible values calculated on the basis of other available information. In this statistic, imputation is performed by copying the most recent report from the unit that has not yet reported. This is usually the previous year, but it may also be an older report if data are missing for several years. For new respondents who have not reported at all, no imputation is performed. Consequently, these are not covered, as there is also no grossing up. This has no significant impact on the overall figures, as new respondents are small.

Missing quantity information is estimated based on reports for the same commodity code from other respondents in the same quarter. If there are insufficient reports with quantity information for a given commodity code, no quantity estimation is carried out. In such cases, the quantity information will be missing in the published figures.

An extended quantity estimation is carried out when data on Prodcom codes are compiled for Eurostat. In this case, all missing quantity information is estimated, incorporating data from external trade in goods and from other EU countries.

There is no grossing up.

Adjustment

No adjustments are made for this annual statistic.