Hungary - Real Retail Sales

The Retail Trade Turnover Indices are business cycle indicators which show the monthly activity of the retail sector in value and volume. The volume measure of the retail trade turnover index is more commonly referred to as the index of the volume of (retail) sales.

Retail trade turnover indices are short-term indicators for final domestic demand.

It is the objective of the turnover index to show the development of the market for goods and services. Turnover comprises the totals invoiced by the observation unit during the reference period, and this corresponds to market sales of goods or services supplied to third parties. Turnover also includes all other charges (transport, packaging, etc.) passed on to the customer, even if these charges are listed separately in the invoice.
Turnover excludes VAT and other similar deductible taxes directly linked to turnover as well as all duties and taxes on the goods or services invoiced by the unit.
The indices of domestic and non-domestic turnover require turnover to be split according to the first destination of the product based on the change of ownership. The destination is determined by the residency of the third party that purchased the goods and services. Non-domestic turnover is further sub-divided into turnover despatched to euro-zone countries and all other non-domestic turnover.

Source data

The production of indices within Member States is normally based on the compilation of data from numerous sources. Detailed methodological information about Member States practice is available from the STS sources.

Statistical surveys

All national statistical authorities have statistical questionnaires used for compiling STS. However, their content and style vary greatly, partly because of cultural differences and partly because of the greater or lesser importance attached to respondent burden and cost. These influences as well as others determine what information the national statistical authorities think that they can observe. In most of the national statistical authorities, the surveys are rarely restricted to one standard questionnaire or form but tend to be a combination of forms, differentiated by major characteristics, namely:

• the activity, size, legal form and the type of variables asked on the form (output, prices, employment, other specialised variables);
• occasionally, an extra characteristic, the geographical location of the unit, may influence the contents of a survey.

Administrative sources / registers / declarations

For the purposes of business statistics a limited definition of administrative sources can be used. According to the purpose they serve, administrative registers can be subdivided into basic registers and specialised registers. Examples of indicators which use more frequently administrative sources are turnover (VAT declarations) or number of persons employed.

Estimations

The STS-Regulations explicitly permit the use of statistical estimation procedures. For example, these may be used for item or unit non-response, grossing of sample results to the level of the frame population or to adjust results from surveys or administrative sources where the frame population does not match sufficiently the target population or the variables collected are not sufficiently close to those required. Hence, this need for estimation may arise because of non-response or because the statistical authority has chosen not to collect directly the information required.

Non-official sources

There is a great variety of non-official data, much of it available from consultancies or research institutes. Trade associations and chambers of commerce also produce non-official data about the business community. With only a few exceptions, private research institutions do not carry out regular surveys and tend to produce results from ad hoc surveys for clients.

Data collection

Detailed methodological information about Member States practice is available from the STS sources.

Data validation

Data received from the countries and the European indices are validated using logical validation rules.

At national level, editing involves studying data from respondents with the aim of identifying (and eventually correcting) errors. Not all errors can be identified and the aim is to detect the errors that have a significant influence on the results. Rules to assist in identifying errors may flag possible errors that require further investigation to determine where there really is an error as opposed to an unusual result or they may identify definite errors. Editing involves checks for completeness, that values are within given ranges and that values for related variables are coherent. Data editing may take place during or after data entry.

Responses can be compared to the response of previous months. Inconsistency or large deviations (outside of a pre-established range) indicate that a closer look is desirable. This may result in editing. In the context of timeliness, the editing process may be designed to give top priority to those outliers that are most in need of editing for the sake of reliable aggregates. By solving the worst cases, large improvements can be achieved.

Eurostat also carries out validation checks on the national aggregated indices it receives.

Data compilation

National level

The starting point for the processing stage is the information as collected from respondents. The aim is to bring these data to the level of the intended statistical output. For various reasons, the act of processing comprises more than just aggregating questionnaire items.

Processing steps can be summarised as follows.

• After data entry, errors and inconsistencies are detected and corrected during editing.
• Subsequently, item non-responses as well as gaps between questionnaire concepts and output concepts are dealt with by imputation.
• The resulting set of clean and complete micro data serves as the basis for weighting and reweighting. During this stage, also frame errors may be accounted for.
• The aggregated data may then be confronted with related data from other sources and possibly integrated.
• Finally, where appropriate, statistical compilations and analysis are carried out, resulting in a non-public data set. Prior to dissemination, the one remaining stage is to identify and treat confidentiality.

European level

The European indices are calculated from national indices, taking into account the relative share of each Member State in the appropriate geographical aggregate, for the gross and working day adjusted forms. This is done at each level of the activity classification level. Only after calculation at all levels of classifications are the EU indices analysed to produce seasonally adjusted and trend series.

However, the data received from each country may need a certain amount of pre-treatment before the EU indices can be calculated. Three necessary stages can be identified as well as one extra stage that is not directly needed for the calculation of EU indices.

Firstly, any data supplied in absolute figures need to be compiled as indices. Secondly, base years need to be harmonised. Thirdly missing activity aggregates need to be calculated. Finally, any of the required forms (for example seasonally adjusted) that are missing are produced, although these are not used for compiling geographical aggregates.

The procedures for compiling the geographical aggregation starts with the gross and working day adjusted series. The European aggregates start - with any number of countries - from the reference period for which 60% of the total weight is reached; as new series pile up, the total weight increases, to reach eventually 100% of the target aggregate. Thresholds also apply to the ending portion of the series; missing countries are approximated by ARIMA (autoregressive integrated moving average) forecasts. More information on ARIMA models can be found at http://www.bde.es/servicio/software/econome.htm).

The weighting system used by Eurostat plays a double role, to carry out geographical aggregation and, when national statistical authorities choose not to provide higher levels of activity classifications, to make activity aggregation as well. The current weighting system uses 2005 data. The weights are sometimes confidential, especially at a detailed level. This can be because the weights are in general based on SBS data which itself may be confidential. The tables containing non-confidential weights can be found here.

Data are revised when additional information from national statistical authorities becomes available. Major revisions and changes in methodology are announced in the monthly News Releases and/or in the publication Quarterly Panorama of European Business.