Which formats are available
The following formats are available when downloading the data
- spreadsheet (.xlsx)
- text (.tsv)
- SDMX-CSV
- SDMX-ML
- JSON-STAT
Spreadsheet (.xlsx)
This is a format based on XML that can be used by spreadsheet applications, such as Excel. The data are stored in a text file that uses XML to define all its parameters. This format is only available for datasets whose size does not exceed 750 000 cells.
Text (.tsv or .cvs)
TSV stands for Tab Separated Value. TSV file is a flat file, which uses the tab character to delimit data and reports 1 time series per line.
CSV stands for Comma Separated Value. CSV file is a flat file, which uses the comma (,) character to delimit data and reports 1 observation per line.
Both TSV and CSV files can be created and modified using spreadsheet software, like Excel.
Here is an example of a TSV data file for a dataset with a single time series:
| FREQ,CURRENCY,POST,STK_FLOW,S_ADJ,PARTNER,GEO/TIME_PERIOD | 2024-01 | 2024-02 | 2024-03 | 2024-04 | 2024-05 |
|---|---|---|---|---|---|
| M,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | 100001 | 96664 | 114413 | 107945 | 107837 |
Here is an example of a data file for a dataset with multiple time series:
| FREQ,CURRENCY,POST,STK_FLOW,S_ADJ,PARTNER,GEO/TIME_PERIOD | 2024 | 2024-01 | 2024-02 | 2024-03 | 2024-04 |
|---|---|---|---|---|---|
| A,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | 1353651 | : | : | : | : |
| M,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | : | 100001 | 96664 | 114413 | 107945 |
In the first column, the header line is written in upper case letters. The sequence of codes is separated by a comma, followed by a code separated by a back slash (\).
Dimension identifiers
The codes separated by a comma (,) are the identifiers of the dimensions. These are used to identify time series.
- header line: the letter in the column FREQ represents the frequency
- time periods are sorted in ascending order, the notation for time period code is:
- year: YYYY (example: 2018)
- semester: YYYY-SN (example: 2015-S1)
- quarter: YYYY-QN (example: 2015-Q1)
- month: YYYY-NN (example: 2015-01)
- colons: a cell contains a colon (:) if no data value is available
- all columns except on the first line contain either values or colons
- flags are attached to values when they are available. The separator used between a value and a flag is a blank.
- the decimal symbol used is the dot (.)
- With a multiple time series:
- if there are 2 time frequencies, such as A (=annual) and M (=monthly), then there are 2 distinct time series: 1 time series for the annual data (FREQ = A) and 1 time series for the monthly data (FREQ = M)
- for a time period that does not match the frequency indicated in the series key, the colon means not applicable. As an example, if the series key indicates A (=annual) as a frequency, there cannot be a value for a monthly time periods. Here is an example with multiple time series:
| FREQ,CURRENCY,POST,STK_FLOW,S_ADJ,PARTNER,GEO/TIME_PERIOD | 2024 | 2024-01 |
|---|---|---|
| A,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | 135000 | : |
| M,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | : | 4700 |
is to be understood as
| FREQ,CURRENCY,POST,STK_FLOW,S_ADJ,PARTNER,GEO/TIME_PERIOD | 2024 | 2024-01 |
|---|---|---|
| A,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | 135000 | not applicable |
| M,MIO_EUR,010,CRE,NSA,EXT_EA16,EA16 | not applicable | 4700 |
SDMX
The Statistical Data and Metadata Exchange (SDMX) is an initiative designed to manage and automate the process of data and metadata exchange. It is a business choice, as opposed to a technical one, that aims to improve the quality of exchanges through standardisation, automation, validation, and data sharing.
This standard describes and universalises the way to exchange statistical data, and provides standard formats for data and metadata, content guidelines as well as IT architecture for exchange of data and metadata. The SDMX message formats have two basic expressions, SDMX-ML (using XML syntax) and SDMX-CSV.
JSON-STAT
JSON-stat is a simple lightweight JSON dissemination format best suited for data visualisation, mobile apps or open data initiatives. For more information about JSON-stat please visit the json-stat website.