Statistics Data Warehouse help - Data Discovery

Browsing data

Using the Data Navigation Menu

You can browse through SDW data using the following navigation hierarchies. Have a look on the right side, to see the main areas of the current navigation menu of the SDW.

Publications Browse data by a variety of pre-defined SDW reports. View online versions of regular statistical publications (e.g. Statistics Bulletin) and retrieve latest values of indicators published in a selected report via the data part Navigation menu
Full Content With the focus on the available datasets that the SDW offers, this section provides metadata information, browse and export options for each dataset
Economic Concepts Browse data by a variety of economic concepts such as prices and costs, financial markets and labour markets
My Settings Customize the user profile, change the options for displaying the data, adjust the sorting and filtering settings, based on your expertise and preferences, and alter the default search
Help All information about the underlying data model, setting up your profile and the usage of different concepts and functionalities in SDW like browsing, searching, filtering, selecting & downloading data, web services
Login The link for the user authentication, a must-do for getting access to the personal reports and data groups
Currency Converter An utility to convert specific amounts from one currency to another at specific dates
Site Directory Offers the big picture of the navigation menu, including a search feature for a smooth exploration of the report and data knots
Contact The contact point for feedback or any additional information related to the SDW
Glossary A dictionary of statistical and financial concepts.
Disclaimer Notes regarding the copyright and the use of logos or data content.
My Publications Compile, store and manage selected Publications pages for future use. My Publications groups are private. Adding, editing, deleting or viewing the personal publications requires a user login.
My Data Groups Browse data previously selected and stored by you as private data groups. This requires you to log in to your SDW account. You can also subscribe to Notification Service and be alerted each time the data are updated.

To begin browsing SDW data, select one of the navigation options from the left-hand navigation menu which is expanded with every click, showing a list of items. The vast majority of the "second-level" items contain further hierarchical levels, which can be expanded in the same manner. Items may contain different numbers of hierarchical levels, depending on the breadth, content and amount of data associated with them.

With every click on a parent node, the page offers more information regarding the section you have just browsed. For example by browsing the "Money, credit and banking" node, the content of the page will list a short description about the data, together with all/part of the information described below:

  • Overview: a brief table of content for the browsed node, offering quicker access to the lower hierarchical levels.
  • Datasets: a list of all datasets contained within the current hierarchy, whose data can be directly browsed, exported, downloaded from here.
  • Publications and Dashboards: all the publications related to the current browsed data.
  • Methodology: additional documentation concerning regulations, explanatory notes, data collection.

image N/A image N/A image N/A image N/A

A navigation item preceded by a plus sign ( + ) indicates that the item contains further sub-items and can be expanded. A navigation item preceded by a minus sign ( - ) indicates that you are currently browsing within that item. The most recently selected item is highlighted in blue. As you browse through the hierarchy you will see that all relevant information (metadata) pertaining to your selections appears within the main content area. Navigation items not preceded by a ( + ) or ( - ) sign are at the lowest level of the hierarchy (leaf nodes). Selecting one of these items will, in most cases, display the Data Selection page, which will present the data pertaining to your hierarchy selection and allow you to select data at a more granular level.

Searching for data

Please note that you are able to customize the search mechanism via the My Settings page located on the left-side menu. To begin searching for data, enter your search criteria into a text box located in the top right-hand corner of each SDW page and click on the Search button.

Google search Search for free text via Google.
SDW search Search for free text or series keys via the internal search mechanism.

Submitting Series or Metadata Search

The Series Search is the default SDW search option. If you want to search among the metadata please select the ‘Metadata search’ radio button.

Enter your search criteria into the search text box and click on the Search button. A message will appear informing you that your search is being conducted.

Please keep in mind that there are datasets loaded to SDW from many different data sources. The terminology used for the series title might differ. The SDW search engine handles well the general grammatical issues, like nouns in singular/plural, verbs in different tenses, names of countries and their adjectives, but unusual abbreviations are not handled automatically: e.g. the following two search terms will search for the same series French small enterprise loans and small enterprises France loans. In addition, euro area and euro zone are considered equivalent therefore using any of the 2 terms in the search criteria the returned results should be the same. For example, euro area unemployment will return the same results as euro zone unemployment.

Note that the free text search will, by default, retrieve series whose description or short title includes your search terms.

If you want to search only series from a specific dataset, it is enough to add the dataset name to the search criteria like in the following example: BSI euro area

If you want to search only series from a specific Publication, it is enough to add the dataset name to the search criteria like in the following example: Statistics Bulletin euro area.

However, combining two or more search terms with the OR logical operator requires explicit definition - e.g. population (Germany OR France). Logical operators ( AND , OR and NOT ) must be specified using capital letters (please see the data grid below for all the available operators).

The order of search terms is unimportant. Therefore, entering seasonally adjusted unemployment Luxembourg and Luxembourg unemployment seasonally adjusted would yield identical search results. To search for a particular phrase, simply enter the phrase in double quotation marks- e.g. "gross domestic product". In this case, the order of the search terms is important.

You can also search for time series by series key. Enter your series key search criteria into the search text box and click on the Search button. A message will appear informing you that your search is being conducted.

You may enter partial or complete series keys/aliases with the option of using wildcards for series dimension values. The following wildcards are permitted:

  • ? or * for multiple characters (e.g., BOP.M.? or EXR.*.EUR.* ),
  • ^ for a single character (e.g., EXR.^.USD.EUR.SP00.A )

You may also enter a list of complete series keys/aliases separated by a comma, a tabulator or a blank space. In the same way, you may search for multiple wildcarded series keys separated by a blank space, as below.

  • BKN.H.U2.*.ZZ.S.Q BKN.H.*.A020.Z.ZZZZ.ZZ.S.Q EXR.A.*

The length of time between the submission of your search criteria and the appearance of the search results will depend on the breadth (specificity) of your search criteria. Broader searches that specify very general terms (for example, GDP , HICP or M3) typically take longer to process than more restrictive searches (such as GDP euro area or HICP euro area energy ), and are likely to result in thousands of series whose titles contain these general terms. Restricting your search criteria will usually result in faster processing and more relevant search results. Please note that search performance may also be affected by network bandwidth availability .

The following text search operators are available:

Name Operator Description
And (default) AND Use the AND operator to search for data items containing all search terms specified. However, since this is the default logical operator, it is not necessary to explicitly include AND between search terms.
Or OR Use the OR operator to search for data items containing any of the search terms specified.
For example: population (Germany OR France), which will return data items containing either German or French population data.
NOT ~ Use the NOT operator ( ~ ) to search for data items containing one search term and not the other.
For example: "euro area" M3 ~ "Index of Notional Stocks", which will return all data items that include M3 and euro area but do not include Index of Notional Stocks.
Soundex ! Use the soundex ( ! ) operator to expand a search to include words that have similar sounds, i.e. words that sound like other words. This function allows comparison of words that are spelled differently but sound alike in English.
For example: !labour productivity, which will return data items that include productivity and labour or labor.
Stem (default) $ Use the stem ( $ ) operator to search for terms that have the same linguistic root as the search term. (This operator is applied to each search term by default. No need to add it)
For example: "euro area" $employ, which will return data items containing euro area and employment, employed, or employees.
Exact search ] Switches OFF the default stem ( $ ) operator, and the search only returns terms which fit exactly the search term prefixed with ].
For example: "euro area" ]employment, will return time series with a description or title attribute containing euro area and employment, but it will not return results with employed, or employees.
Wildcards %
Wildcard characters can be used in search expressions to expand word searches into pattern searches.
The percent ( % ) wildcard denotes a string of characters appearing in positions indicated by the wildcard. For example: %euro employ%.
The underscore wildcard ( _ ) denotes the occurrence of any character appearing in the position indicated by the underscore. Right truncation involves placing the wildcard on the right-hand side of the search string. For example: "euro area" M_, which will return data items containing euro area and M0, M1, M2, etc.
Left truncation involves placing the wildcard on the left-hand side of the search string. The left-truncated and right-truncated searches may be combined to create double-truncated searches.
Parentheses (
Parentheses are useful for defining search criteria in a more concise manner, or in constructing complex searches that involve multiple search operators. For example:
  • "euro area" (GDP OR M3), which will return data items containing either euro area GDP or euro area M3.
  • "euro area" (GDP OR HICP) NOT cyclical% , which will return data items containing either euro area GDP or euro area HICP, but without phrases beginning with cyclical.
Parentheses are also useful when specifying search terms to be excluded. To exclude more than one term from a search, combine the terms using the OR operator. For example: euro area" HICP NOT (energy OR food).
To exclude more than one term only if they appear together, combine the terms to be excluded using the AND operator. For example: "euro area" HICP NOT (energy AND food)

Search Results

The Search Results page appears once your search is complete. After submitting the search, you end up on the Data Selection page which contains three different areas - Available Datasets, Dimension filters and List of series (not all of which will appear for every search, as this depends on the type of search conducted and the nature of the search results).

For the Series Search, the following areas are displayed:

Reference Series A list of reference series matching your search criteria. The number of reference series displayed appears in parentheses. Not all search results will contain reference series.
Published Series A list of published series (series published- e.g. in the Statistics Bulletin or ESRB Risk Dashboard) matching your search criteria. The number of published series displayed appears in parentheses. Not all search results will contain published series.
Series A list of time series sets matching your search criteria. The number of time series sets returned appears in parentheses.

The Search Result page contains for the Metadata Search:

Explanatory items A list of datasets of which the explanatory notes (metadata) match your search criteria (text search only). The number of items displayed appears in parentheses.

Please note that the maximum of displayable series on Search Result page is 2000 per dataset (maximum 2 data cubes of up to 1000 series each). If your search criteria result in more series, only the first 2000 will be displayed and you will be asked to refine your query.

Your search criteria are displayed at the top of the page, along with information on the number of series, siblings and explanatory items retrieved as a result of your search. For quick access to these areas, click on one of the numbers displayed. Each area may be collapsed and expanded as needed using the icon show / icon hide icons displayed in the top right-hand corner of the area.

Each series title is a hyperlink on the search result page of the Series Search. It leads to a Quick View page.

Filtering & Selecting data

The Data Selection page is the primary "launch pad" for performing many key SDW functions. It is usually presented as a result of:

  • browsing for data through the left-hand SDW navigation menu;
  • making a request to browse an SDW search results selection;
  • accessing a data group through My Data Groups.

Data appearing on the Data Selection page may be:

  • filtered: data may be filtered on the basis of common dataset dimensions. Note: In order to see a list of "Selected Series" relating to your selection, your filtering criteria must be less than 500 series. To make a more detailed selection you can use the "Base Filters", such as "Frequency" or "Geographic Area", or you can go to the "Advanced Filters" section where you can apply filters per each dataset. You may also deselect some of the "Available Datasets" which appear at the top of each "Data Selection" page.
  • viewed: selected data may be viewed and printed in tabular or chart format. Metadata associated with the selected data may also be viewed.
  • exported: selected data may be exported in CSV & SDMX-ML version 2.0 Compact format.
  • stored: selected data may be compiled and stored on a long-term basis as a private or public data group. This data may be accessed and managed through the My Data Groups option.

The Data Selection page contains three areas:

Available Datasets A list of datasets retrieved via data browsing or a data search, constrained by any filter criteria selected. The number of datasets displayed appears in parentheses.
Dimension Filters Options for filtering the retrieved time series on the basis of common dataset dimensions.
Selected Series A list of time series retrieved via data browsing or a data search, constrained by the filter options. The number of time series presented appears in parentheses. When only one dataset is selected, the Common Description of the series is displayed above the list of series. This contains the description of all shared dimension codes between the available series. Changes in the Dimension Filters update the Common Description too.

On a Data Selection page, the relevant button inside the "data functions toolbar" is highlighted in blue:

Screen data table

Click on a button from the toolbar to view Data Table, Data Chart or Metadata, to apply Advanced Filters criteria, or to Data Download or save (Add to My Data) the time series of your choice.

Available Datasets

The Available Datasets area displays a list of datasets that are associated with time series retrieved via data browsing or a data search, constrained by any filter criteria selected. Each dataset displayed is therefore associated with at least one time series in the Selected Series area, if no filters have been set. The number of time series retrieved for each dataset appears in parentheses next to the dataset name. All datasets displayed here are selected by default. To unselect a dataset (if there is more than one dataset present), uncheck the corresponding checkbox. All series associated with the dataset will be removed from the Selected Series area. To re-select the dataset, click on the corresponding checkbox.

The "Available Datasets" and "Dimension Filters" areas can be expanded/hidden as per user needs. For this purpose, use one of the icons below.

hide icon / show icon Collapse/Expand area - click on this to hide/show the Dimension Filters or the Available Datasets areas.

Two different MS ExcelPivot export options are available next to each dataset:

pivotx icon Export Data into standard MS Excel Pivot Table - click on this to export the dataset to a static Excel workbook. Data are displayed similarly to data exported using the Refreshable Pivot Table, but in this case to get the latest values available, the data need to be exported again from the SDW portal.
(After MS Excel has opened the exported Excel workbook, please ‘Enable Editing’ and ‘Enable Content’ in the workbook.)
pivotx icon There is also a refreshable MS Excel Pivot based on the SDMX 2.1 Web Services as data source. Please download the Excel workbook, open it, and start downloading the data you earlier selected on the SDW portal. The data will be downloaded to a worksheet, and the pivot table will be created on a separate worksheet. The data worksheet can be refreshed any time, which will also trigger the automatic update of the pivot table.

Dimension Filters

The Dimension Filters offer a mechanism for filtering time series on the basis of common dataset dimensions. You can use the Dimension Filters at any time to apply further data restrictions and refine your current selection, regardless of how you arrived at the Data Selection page (i.e., through data browsing, a data search, etc.).

The Dimension Filters contains the key dataset dimensions shared by the series appearing in the Selected Series area. Here you can apply filters that are common to all selected datasets but you can also refine the results by applying filters on each dataset. The "Frequency" and "Geographic Area" filter dimensions are almost always both available, since these dimensions are common to most datasets. Note that more options are available in the Advanced Filters which allow you to filter the data by specific dimensions for each of the selected datasets. Filter dimensions that contain only a single value are "greyed out". Bear in mind that you can customize the way filters behave on the My Settings page.

Filter dimension values are often interdependent and are adjusted dynamically according to your filter criteria selections. If, for example, you modify the Frequency filter dimension from Monthly to Quarterly, the filter option values for the other common dimensions may change to reflect only the available values for quarterly data. Note that each time you select or change a value from a filter dimension, the entire Data Selection page will reload with the updated filter restrictions applied to the data, as well as a refreshed set of filter option values "tailored" to your current filter criteria. This mechanism prevents you from specifying a filter combination that does not yield any data.

To filter data, select the desired filter criteria from the dimensions available. To select multiple values for a particular dimension, click on the filter and pick codes from the list with your mouse one by one.

It is also possible to filter the long code lists. Start typing in the selected filter, and the codelist will be restricted to entries matching your search term. See the illustration below.

SDW Long list codes

The code of any criteria selected within a filter always appear at the top of the list. When you move your mouse over the code, the description of the code becomes visible in the tooltip.

You can always delete an applied filter by clicking the "x" sign next to the filter code. In case you want to remove all the filters that have been applied, please click the "Reset Settings" button.

Advanced Filters

Here you will find additional options to further filter your selection.

The Advanced Filters operate exactly like the primary filter options. Beforehand you need to select the dataset for which you would like to apply more filters (click the dataset within the Advanced Filters) and then select one or more values for the targeted dimension and click out of the field. Each time you perform a selection, the page will reload with the updated filter restrictions applied to the data on the Data Selection page, as well as with a refreshed set of filter option values "tailored" to your current filter criteria. The number of values available for each dimension are written in parantheses next to the dimension name. You are able to customize the settings for filters at any time by accesing the My Settings page. Also notice that the "Reset Settings" button will clear out all the applied filters.

Common Description

The Common Description area displays all filter criteria applying to your selection. It always appears when your data selection pertains to datasets that share the same dimensional structure. It indicates the dimension names as well as the filter values applied to the dimensions. Together with a short series title, displayed in the Selected Series area, it forms a complete description of the series.

Selected Series

The Selected Series area displays a list of time series retrieved via data browsing or a data search, constrained by any filter criteria selected. If a selection made from the left-hand navigation menu or from the search contains more than 500 time series (which is often the case), you will be asked to restrict your search to reduce the number of series presented. Each series is identified with a title, series key, and begin/end dates.

Series are displayed in different shades of blue and grouped in the following order:

  1. Reference Series;
  2. Published Series - series that are published, e.g. in the Statistics Bulletin or ESRB Risk Dashboard;
  3. Other Series - non-reference series that are not published.

The vast majority of reference series are also published. Note that not all navigation items and search results will contain reference and published series.

To select or unselect a series, check the corresponding series checkbox.

Series history

The following icons appears next to each series key:

vintage Series Dataset Vintage – click on this to view the history of a particular time series taking into consideration the update dates at dataset level. The Dataset Vintage page displays the series history organised by the dates on which any series in the dataset has been updated (horizontal time axis), no matter if the selected series has been updated or not at that specific point in time. All series of a dataset have the same horizontal time scale of changes. Changes in observation values belonging to the same period are marked in bold. The deletion of observation values are indicated by Values deleted icons. Use the Export button to download the Data Vintage page into a CSV file. Please note that for now the series vintages page is available only for the Real Time Database (RTD) data.

SDW help © 2015 ECB