A New Partnership

Continuum is proud to announce a new partnership with Thomson Reuters. Thomson Reuters QA Direct is a financial dataset that offers a wide range of high-quality financial metrics. These datasets, centered around quantitative equity management, are rich with financial information reported around the world. Continuum and Thomson Reuters have partnered to bring you this dataset on Continuum’s premier hosted analysis platform: Wakari.

A Shift

Thomson Reuters has done significant work in maintaining and curating their commercial data, especially in the face of a drastically changing economic world: company creation/death/mergers, ticker changes (C-> Citibank to C-> Chrysler and back to C-> Citibank), etc. We at Continuum have also seen shifts in the financial world, namely the inclusion of Python as the language of choice for many in the quantitative world.

One of the reasons that Continuum’s partnership is so fruitful is because we can bring our Python expertise to the Thomson Reuters DBs in order to better enable users to analyze the data. Fund and equity managers, quants, etc. should be focused on analysis and not on the often time-consuming task of writing SQL to extract the data. For this reason, Continuum has wrapped a Python API over Thomson Reuters standard SQL interface. This API delivers the requested data in what is arguably the best object for manipulating time series data: a Pandas Series.

Data Layout

The schema for Thomson Reuters is quite large — one of the reasons why we started writing a Python API. QA Direct contains several databases which store logically separated datasets. Often though, information from several, if not all DBs, will be integrated together for a more complete forecast. We’re starting our integration with a few of these:

  • Worldscope Fundamentals
  • Datastream Equity Pricing
  • I/B/E/S

QA Direct is the metadatabase which maps these disparate databases together. Worldscope contains standardized presentations of publicly reported financial statements (balance sheets, income and cash flow statements) for 57,000 companies globally. Datastream houses historical end of day pricing and dividends, and lastly, I/B/E/S (Institutional Brokers’ Estimate System) stores estimates and recommendations by brokers and analysts.

API Design

We appreciate the skepticism quants bring when approaching new products, and so we are doing two things to alleviate those concerns:

  • The API is open-source and made available in Wakari
  • The API focuses on delivering data rather than baked or half-baked analysis functions
    • We are providing convenience functions to easily plot data with Bokeh, Wakari’s interactive plotting tool
tr.query(security,measurement,frequency)

 

Python

 

 

We prefix the Thomson Reuters id with ws to ask for the appropriate id in the Worldscope DB.

Below are a few examples of how one might pull filing information such as Net Income for Apple (ticker: AAPL). The balance sheets are store in the Worldscope DB.

apple = tr.query('AAPL','ws.1551','Q')

 

Python

 

 

If you wanted to do a deep-dive on a given security, you would want to list all of the metrics like so:

apple = tr.query('AAPL',['ws.1551','ws.2001','ws.3351'],'Q')

 

Python

 

 

Above we are asking for the Net Income, Cash and Short Term Investment, and Total Liabilities. But we also may want to investigate a single metric for a list of securities:

net_income = tr.query(['IBM','AAPL',...],'ws.1551','Q')

 

Python

 

 

And as we mentioned before, plotting is straightforward.

Integrated Plotting With Wakari

apple = tr.query('AAPL',['ws.1551','ws.2001','ws.3351'],'Q')
apple.plot('ws.1551','Apple: Net Income')

 

Python

 

Additionally, we recognize that the biggest bottleneck for performance will be pulling the data from Thomson Reuters DBs. To mitigate the constraints of data non-locality, the API leverages IOPro’s PyODBC module, which is optimized for the ODBC connection to Microsoft SQL Server-based QA Direct. While IOPro is not open-source, it is available to all Wakari users.

Data Play

Both Thomson Reuters and Continuum are ardent supporters of playing with data and as such, Thomson Reuters has generously agreed to provide a teaser dataset which will be limited to 37 securities but will make available all metrics for those securities. The API will be released in the coming weeks. If you already have Thomson Reuters credentials we invite you to log onto Wakari and start using the API right away. If you don’t already have Thomson Reuters credentials and are interested in exploring the data please signup for Wakari and contact Continuum at [email protected] for access to the teaser data set.


About the Author

Ben Zaitlen

Data Scientist

Ben Zaitlen has been with the Anaconda Global Inc. team for over 5 years.

Read more

Join the Disucssion