[go: up one dir, main page]

Menu

#313 API data access

Backlog
open
nobody
None
5low
2025-09-19
2023-12-01
Steve Keen
No

Quick query, inspired by finding out that one of my favorite data sources (the BIS) will no longer make its data available in CSV format from January 2024.

Is it possible to add an import routine that uses a site's API, in a way that doesn't involve the user in effectively having to code their access?

The BIS notes here https://www.bis.org/statistics/sdmx_techspec.htm that they use a standardised model for data handling that is shared across a number of agencies:

The SDMX information model is a standardised object model for modelling statistical domains centring on the structure of the data sets, the coding schemes used for classifying series, and the rules for controlling the exchange of data. The model also allows for the definition of various business rules including those for checking data quality and transforming their structures and classification schemes. Means of categorising and indexing the data are provided to help end users such as subject-matter statisticians discover the datasets and series they need.

I expect that this would be useful to corporates for data access too.

Discussion

  • High Performance Coder

    • Priority: 4Urgent --> 6normal
     
  • High Performance Coder

    This sort of thing is better handled by an auxilliary python utility.

    I suspect, if we remain in this business, that we'll need to provide a curated list of such utilities.

    But this is for later...

     
  • High Performance Coder

    • Milestone: Pascal --> Backlog
     
  • High Performance Coder

    Ticket moved from /p/minsky/ravel/437/

    Can't be converted:

    • _priority: 6normal
     
  • High Performance Coder

    • priority: 6normal --> 5low
     
  • High Performance Coder

    Ticket moved from /p/minsky/tickets/1830/

    Can't be converted:

    • _priority: 5low
     

Log in to post a comment.