Statistical Data and Metadata eXchange (SDMX) in Python¶
sdmx (‘sdmx1’ on PyPI) is a Python implementation of the SDMX 2.1 (ISO 17369:2013) and 3.0 standards for Statistical Data and Metadata eXchange.
The SDMX standards are developed and used by national statistical agencies, central banks, and international organisations.
sdmx can be used to:
Explore and retrieve data available from SDMX-REST web services operated by providers including the World Bank, International Monetary Fund, Eurostat, OECD, and United Nations;
Read and write data and metadata in file formats including SDMX-ML (XML), SDMX-JSON, and SDMX-CSV;
Convert data and metadata into pandas objects, for use with the analysis, plotting, and other tools in the Python data science ecosystem;
Apply the SDMX information model (IM) to structure and publish your own data;
…and much more.
Get started¶
The SDMX standards are designed to be flexible enough to accommodate nearly any data. To do this, they include a large number of abstract concepts for describing data, metadata, and their structure and relationships. These are collectively called the information model (IM) of SDMX.
The documentation you are reading does not (and could not) repeat full descriptions of the IM or other parts of the SDMX standards.
Instead, it focuses on the precise, Pythonic, and useful implementation that this sdmx package aims to provide.
Detailed knowledge of the IM is not needed to use sdmx; see a usage example in only 10 lines of code or a longer, narrative walkthrough.
To learn about SDMX in more detail, use the list of resources and references, or read the API documentation in detail.
sdmx user guide¶
- Data sources
- Data source limitations
ABS: Australian Bureau of Statistics (SDMX-ML)ABS_JSON: Australian Bureau of Statistics (SDMX-JSON)AR1: National Institute of Statistics and Censuses (Argentina)BBK: German Federal BankBIS: Bank for International SettlementsECB: European Central BankESTAT: Eurostat and relatedILO: International Labour Organization- International Monetary Fund
INEGI: National Institute of Statistics and Geography (Mexico)INSEE: National Institute of Statistics and Economic Studies (France)ISTAT: National Institute of Statistics (Italy)LSD: National Institute of Statistics (Lithuania)NB: Norges Bank (Norway)NBB: National Bank of Belgium (Belgium)- Organisation for Economic Cooperation and Development
SGR,SGR3: SDMX Global RegistrySPC: Pacific Data Hub DotStat by the Pacific Community (SPC)STAT_EE: Statistics Estonia (Estonia)StatCan: Statistics CanadaUNESCO: UN Educational, Scientific and Cultural OrganizationUNICEF: UN Children’s FundUNSD: United Nations Statistics DivisionUY110: Labour Market Information System (Uruguay)WB: World Bank Group “World Integrated Trade Solution”WB_WDI: World Bank Group “World Development Indicators”- Source API
- API reference
- Implementation notes
- How to…
- What’s new?
- Next release
- v2.27.0 (2026-08-06)
- v2.26.0 (2026-04-04)
- v2.25.1 (2026-01-23)
- v2.25.0 (2025-12-24)
- v2.24.0 (2025-12-18)
- v2.23.1 (2025-10-01)
- v2.23.0 (2025-09-30)
- v2.22.0 (2025-03-25)
- v2.21.1 (2025-01-14)
- v2.21.0 (2025-01-13)
- v2.20.0 (2024-12-16)
- v2.19.1 (2024-10-23)
- v2.19.0 (2024-10-23)
- v2.18.0 (2024-10-15)
- v2.17.0 (2024-09-03)
- v2.16.0 (2024-08-16)
- v2.15.0 (2024-04-28)
- v2.14.0 (2024-02-20)
- v2.13.1 (2024-01-24)
- v2.13.0 (2024-01-23)
- v2.12.1 (2023-12-20)
- v2.12.0 (2023-10-11)
- v2.11.0 (2023-08-04)
- v2.10.0 (2023-05-20)
- v2.9.0 (2023-04-30)
- v2.8.0 (2023-03-31)
- v2.7.1 (2023-03-09)
- v2.7.0 (2022-11-14)
- v2.6.3 (2022-09-29)
- v2.6.2 (2022-01-11)
- v2.6.1 (2021-07-27)
- v2.6.0 (2021-07-11)
- v2.5.0 (2021-06-27)
- v2.4.1 (2021-04-12)
- v2.4.0 (2021-03-28)
- v2.3.0 (2021-03-10)
- v2.2.1 (2021-02-27)
- v2.2.0 (2021-02-26)
- v2.1.0 (2021-02-22)
- v2.0.1 (2021-01-31)
- v2.0.0 (2021-01-26)
- v1.7 and earlier
- pandaSDMX (versions 0.9 and earlier)
- Development
Contributing and getting help¶
Read
CONTRIBUTING.mdin the source distribution or in the Git repository (here) and/or Development linked above.Report issues, suggest features, or browse the source code on GitHub.
Join the
#sdmx1Internet Relay Chat (IRC) channel on irc.libera.chat.You can join in a web browser using web.libera.chat, or using any IRC client; see this guide for more information.
Older fora may have answers to some questions:
On Stack Overflow, questions and answers with the tags
[sdmx] [python].On Google Groups, the sdmx-python group.
Neither of these are actively monitored by
sdmxdevelopers.