You are here:  Home  >  Data Blogs | Information From Enterprise Leaders  >  Current Article

Quant Finance Shops Can Add Sentiment About Macroeconomic and Geopolitical Events To Their Rules Toolbelts

By   /  September 28, 2012  /  No Comments

Real-time macroeconomic and geopolitical events and sentiment about them now figure into RavenPack News Analytics 3.0, to help financial firms react more quickly to what’s happening in the world.

The solution is aimed at quant finance shops, such as hedge funds, banks, and some financial research houses, where the machines are doing the trading. It’s a part of the market that’s had some rough times of late because of some trading errors, but in general quant firms represent a growing piece of the financial space, says RavenPack Head of Sales and Business Development Hugh Taggart.

The company’s previous versions have been able to produce the raw data as to the health of 30,000 global companies by weighing the balance of sentiment in articles published online by Dow Jones’ news sources, such as the Wall Street Journal and Barrons’. V. 3.0 now is able to extend that to analyzing macroeconomic events and geopolitical news data. Including stock as well as macroeconomic and geopolitical events “means people who trade things other than stocks – trading foreign exchange, bonds, commodities, any asset class—can now use the data it provides in their algorithms,” Taggart says.

In every news story, RavenPack News Analytics looks for multiple things. It identifies entities, be they companies, places or tradeable assets in a point-in-time sensitive way to account for changing corporate status due to events like mergers; produces a relevance score to tell clients how relevant the story itself is to the entities detected in it and to filter out potential noise; and provides a novelty metric as to whether it is breaking news or a follow-up, as the former tends to be the occasion to get ahead of financial transactions.

Another area is events – most of these are common corporate actions like earnings announcements, but the latest version adds tracking of hundreds of macroeconomic, market-moving events such as terrorist attacks, natural disasters, or federal reserve governor speeches, in real-time. In total over 1,200 macroeconomic, geopolitical, and corporate events are systematically detected and the analysis delivered as a real time data feed, the company says.

Once it has identified the event and location, it goes to work on sentiment. It uses both natural language processing and, as a training set for its algorithms, an expert consensus technique, where analysts did human-level interpretations of sentiment to entities over a thousand articles. That results in an Events Sentiment Score (ESS) based on events specific to an entity. In practice, if reports just surfaced of an earthquake that occurred in Christchurch, New Zealand, it would identify the event and where it took place, give the report a high novelty score and a high relevance score for the earthquake taking place in the location of the identified entity.

“Earthquakes generally are regarded as negative to the financial community,” says Taggart, and News Analytics rounds that out by providing further evident to modify the scores – its Richter scale measurement, for instance, or the number of casualties. “That lets us produce pretty accurate sentiment scores on stories,” he says.

Clients make up their own rules as to how they want to use the data produced. In the above instance, for example, clients trading the New Zealand dollar, could come up with a hypothesis that says, in the instance of a negative geopolitical event, stop buying the dollar. They could gauge those actions according to whether the sentiment is ranked 0, as the most negative, or 100 as the most positive – automatically take a severe action if the sentiment is ranked 25 or below, or send an alert for review if it’s ranked between 25 and 75. Alternatively, instead of event trading, quant firms could take an aggregate of sentiment scores on other macroeconomic or political news to build indices that dictate taking actions when certain extremes are met.

“There are many ways clients could use it but in its simplest form it lets them react quicker to events,” says Taggart. “How they act on it, what they feel that index tells them, is their edge in the market.”



About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

The Future of Big Data: Next-Generation Database Management Systems

Read More →