Quandl Founder Tammer Kamel wants to create a new Wikipedia for Numeric Data. Quandl has built a sort of "universal data parser" which has thus far parsed about 2.8 million datasets without anything from any data publisher. As long as they spit out data somehow (excel, text file, blog post, XML, API, etc) the "Q-bot" can slurp it up. The result is quandl.com as sort of "search engine" for numerical data. Quandl's bot returns data in a totally standard format. The idea with Quandl is that you can find data fast, and it is ready to use.
Quandl has indexed 2 million financial and economic time-series datasets, including 85 global stock market indexes; spot and futures prices for 70 commodities and 10 commodities indexes; 60 futures contracts with prices, commitments, and historical and continuous contract data; exchange rates against US dollar for 200 currencies; and global overviews which provide snapshot data for key stock indexes, commodities prices, interest rates and exchange rates.
Quandl has also enabled datasets to be downloaded directly using the popular R language of statistics. It also enables users to create supersets (or mashups for data).
When you grab one or more columns from various datasets and combine them into a new dataset, you have created a "Superset":
- Supersets are just datasets: everything you can do on Quandl with a dataset, you can do with the new superset you have just created: trim, transform, share, download, API, etc.
- Supersets are permanent. Once you make a superset, it exists perpetually at the same URL until you delete it. And every time you come back to a superset, Quandl refreshes it with the latest data from its source(s).
Some of the Topic Nodes that have been created so far:
- Global Asset Markets -historical and current market data on equities, interest rates, exchange rates, commodities, real estate
- Stocks - stock market indexes from 60+ stock markets from around the world
- Commodities - spot and/or futures prices for 76 physical commodities and 12 commodity indexes from around the world.
- Futures - 200+ futures contracts from 10+ exchanges, with data going back as far as the 1950s. Data indexed includes current market prices, commitment of traders, continuous contracts, and historical contracts.
- Macroeconomic Indicators -a variety of reliable and credible sources, including the IMF, the World Bank, the US Federal Reserve, central banks and government statistical agencies around the world
Every single dataset on Quandl is available via the Quandl API, irrespective of where or how or in what format the data was originally published. Anonymous API calls are limited to 10 requests per (UTC) day. Signed up Quandl users get 100 API calls per day but Users can request higher thresholds. Every dataset on Quandl is associated with a persistent and unique URL that allows users and developers to access that data perpetually, in any format they want. Every dataset on Quandl has its own unique Quandl-ID.
- To get all data points for the Prague Stock Index in JSON, do this call:
- Or, if you want the data as csv, do this:
- Registered users should include their auth_token like this:
You can find your auth_token on the Account page, API tab.
- If you just want a few specific days, do this:
- Append any basic API call with parameters to indicate the desired transformation:
- To get the Canadian GDP annual per cent change:
- Append any basic API call with parameters to indicate the desired frequency.
- You can choose the sort order:
- You can use
rows=nto get only the first n rows of your query. Use
rows=1to get the latest observation for any dataset.
Note there are other dataset portals in the market including Infochimps API and Data Market API. There are 346 APIs listed under the financial category in our directory. Will Quandl help the case of users needing fast, open, free numerical data? I guess it depends on traffic volume, user volume and of course the API calls and mashups created by developers?
Free and Open Numeric Data? Just another Quandl API call away!