Published statistical data are data that are disseminated and presented to the public in electronic or printed form.

Projection - Published before the start of the period to which the results relate. Historical data (time series) and models for analysis and design of time series are used for making projections.

Estimates - Published at the very end or immediately after the end of the reference period. The estimate is a combination of historical data with available partial information about the period to which the results relate. As in the case of projections and estimates, the results are presented at the overall level, without more detailed divisions.

Flash estimates - Estimates that are calculated and published as soon as possible after the end of the reference period, using a set of indicators available at a given time. This set is significantly narrower than the one used for final estimates, so it is necessary to use econometric models as well. Rapid estimates are not necessarily the most accurate measure of the observed phenomenon, but they are certainly timely information and show a tendency to occur.

Previous data (First data) - Calculated and published before final results. They are based on a set of available statistical and administrative data sources, which is broader than the set used for rapid estimates. In addition, calculations are more detailed than quick estimates and are performed at lower levels of aggregation. The calculations are subject to corrections after the set of indicators for the final (final) calculations is completed.

Final data - Based on a complete set of indicators, and calculations are performed at the lowest levels of aggregation. They are subject to revisions if there is a change in methodological concepts and definitions, in calculation methods, classifications, legislation, etc. The final data are not subject to ongoing audits. They can be changed only in case of major revisions.

Revised data are the result of audits.

Estimated and previous data are subject to ongoing revisions due to the inflow of new, more complete information.

General Revision Policy defines the following types of revision:

Current revisions are announced simultaneously with new data publishing. Revised data are explained in regular issues, and in methodological notes, described is the reason for the revision. Revised data must be marked in published materials, starting from the period when the revision was performed.  

Мajor revisions are announced in Release Calendar. After huge revisions, together with the revised data, published is also the accompanying documentation, offering the possibility of estimating new time series. Documentation contains all detailed explanations and reasons for performing the revision, estimation of revision influence on aggregates, presenting comparative review of old and new data, explaining length of revised time series and level of revision details.

Non-scheduled revisions are unforeseen changes in published data, which stem from unexpected events. They are performed upon the detection of minor/serious errors in published data/information or upon the detection of a failure in the process of data production, which leads to the occurrence of errors.

All errors in the statistical editions are corrected as soon as possible, with the clearly marked corrections.

In accordance with the Revision policy, planned changes relating to contents, format, methods and classifications, which can influence the statistical results presentation, are also presented in advance within the edition preceding the one that is to be changed.

Following the Law on Official Statistics, SORS undertakes all prescribed administrative, technical and organizational measures, necessary for confidential data protection from illegal access, publishing or using. Data collected in statistical survey and other activities are exclusively used for statistical purposes and are published in aggregated (sum) form.

In accordance with Article 43 of the Law on Official Statistics at the users’ request and on their account, responsible producers of official statistics may provide information and data obtained by special processing, which are not provided for by the Program and the Plan of Official Statistics.

Custom‐designed analyses are produced in accordance with the Rulebook on the method of  using and providing the data produced by the Statistical Office of the Republic of Serbia. The  requested data are provided without jeopardizing the principle of statistical confidentiality  and they have to be representative to the extent as appropriate.