The leading knowledge platform for the financial technology industry
The leading knowledge platform for the financial technology industry

A-Team Insight Blogs

Trading on Unstructured Data 101 at SXSW

SXSW – for South by Southwest – is a massive festival of music, film and web-related technology that just got going here in Austin, TX. I’ve been attending now for a few years for the music shows, where the event got its start, 25 years ago. But nowadays the ‘Interactive’ side is bigger, and this year I was pleasantly surprised to find a session on “Machines Trading Stocks on News” was on the program. So, after my usual Sunday brunch at Mulberry (brioche french toast and a couple of mimosas), I stopped by.

The panel session was moderated by Adam Honore, research director at Aite Group, and included: RavenPack CEO Armando Gonzalez; Jacob Sisk, senior research scientist at Thomson Reuters; and John Kittrell, who is a quantitative analyst at Knightsbridge Asset Management.

Unstructured data encompasses many sources, include textual news feeds, news and company websites, company and personal blogs, social networks and micro-blogging services like Twitter. A number of companies – including Dow Jones, Deutsche Boerse’s Need To Know News unit, RavenPack, Selerity and Thomson Reuters – take this unstructured data and parse it using semantic analysis to create structured feeds with XML tags.

Trading using unstructured data is growing. Aite Group reckon that back in 2008, just two percent of trading firms were involved in it. Today, that estimate has grown to nearly a third of firms are evaluating adopting it.

Collecting and analysing unstructured data allows a firm to determine so-called sentiment as it pertains to a company. Sentiment – as defined by the panel – being “emotional language” relating to an event that drives a stock price either up or down.

Events might be news items related to company results, products, regulatory approvals and credit ratings. Understanding the context in which words are used is important, for example: “900 layoffs at AOL” as a news story headline has very different context, and likely impact, from a blog post that relates “… and layoffs are down in the auto sector.”

And latency also plays a role, such as in relaying data related to economic bulletins from government departments in Washington DC to futures traders in Chicago, where the impact of such news is usually significant, said Honore.

Recent action in that endeavor includes CoreSite’s hosting of collection and relay systems from Need to Know News and Rapidata at its date centre, located just one mile from the United States Department of Labor, Department of Commerce and Department of The Treasury. Also relevant is the Intellifiber Networks low-latency optical network from DC to Chicago.

Along with a screening of Senna, this session was a nice diversion from the music that will take up the rest of my week. Check out my other blog for some day parties I’m hosting, and other coverage. #donttellmybossshethinksimworking.

Related content


Recorded Webinar: Market data management, licensing and administration in the post-Covid environment

Market data administration has always been a challenge. For many firms, keeping tabs on permissioning and entitlements, compliance with licensing agreements, and reconciling all that with increasingly complex invoices requires a significant dedicated resource with a clear understanding of the issues involved. As if that weren’t enough, things got more challenging for these teams in...


Beeks’ Colocation-as-a-Service Offers Agile Test Access to SGX

Beeks Group’s Colocation-as-a-Service (CaaS) collaboration with the Singapore Exchange (SGX), announced last week, offers market participants rapid access to high-speed connectivity without a major financial or contractual commitment. The facility allows firms to test new approaches and products, using on-demand virtual machine (VM) access, then easily shift to a bare-metal production environment should the test...


Data Management Summit USA Virtual

Now in its 11th year, the Data Management Summit USA Virtual explores the shift to the new world where data is redefining the operating model and firms are seeking to unlock value via data transformation projects for enterprise gain and competitive edge.


Entity Data Management Handbook – Third Edition

Welcome to the third edition of the Entity Data Management Handbook which is available for free download. In this updated edition we delve into the role entity data plays in the smooth running of financial institutions and capital markets, the challenges of attaining high quality data, and various aspects, approaches and technologies involved in managing...