Although machine-readable news has been gaining traction as a delivery channel of real-time, low-latency event information, event data is now being used to help firms gain a competitive advantage that increases profitability and helps manage risk. Event data can be defined as highly structured factual information related to breaking events that trading firms can use to make reliable, real-time programmatic decisions. This form of information is structured and distributed more like market data than other forms of real-time information, such as news or even machine-readable news. This has made it easier for firms who are traditionally quantitatively focused and familiar with market data feeds to more rapidly understand and integrate event data as a new real-time factor into their trading strategies.
Built around the real-time discovery and interpretation of events, such as corporate earnings announcements, employment statistics or court rulings, event data systems combine sophisticated real-time search and extraction technology with proven market data distribution technologies to rapidly identify, distill and deliver machine-readable insight to electronic trading and risk-management firms.
In today's automated financial industry, firms need a 360-degree view of the market to gain an information edge over competition. Although market data gives traders an accurate picture of certain market statistics, event data delivers a comprehensive view of the state of current world events that might impact the capital markets. Having the ability to get valuable, market-moving information into traders' hands faster than ever before improves trading strategies and helps firms gain a leg up on their competition. Without the complete view of the state of global events, a financial firm risks being blind-sided by unexpected news that can impact its short-term and long-term trading strategies.
An Electronic History
The electronification of markets was the biggest factor that gave investment firms the ability to begin automatically trading securities based on breaking events. As execution venues, such as the Chicago Board Options Exchange (CBOE), the Chicago Mercantile Exchange (CME) and other markets, shifted from the floor to computerized trading, market participants for the first time were able to quickly execute trades originating from high-speed computer applications that could process news and events as primary real-time inputs.
Because of this market transformation, sophisticated proprietary trading and asset management firms emerged to take greater advantage of the opportunities electronic trading offered, particularly in the area of event-based trading. One of the first types of events that firms began using in their program trading models were U.S. economic events from government organizations such as the Department of Commerce and the U.S. Department of Labor. Many of the floor traders on the execution venues such as the CME would react to these reports manually in the pits. It was only natural to begin trading these types of events using computers when their business changed from floor to electronic trading.
News vendors began meeting this demand by introducing the first machine-readable news feeds that helped firms include the information harvested by journalists in trading models. Vendors started employing new taxonomies and tagging structures coupled with XML technologies. As electronic trading evolved to reach faster speeds, the buy-side began demanding even lower latency, more highly structured information, which drove the evolution from machine-readable news to machine-readable event data. This new data, comprised of occurrences that are ambiguous and harder to quantify, is restructured into factually based event data that traders can intelligently and confidently analyze and feed into trading strategies, giving them the ability to find opportunities where they hadn't existed in the past. The ability to structure the core and contextual elements of a breaking event sets event data apart from news or even machine-readable news. One of the biggest limitations algorithmic trading models have versus human traders is the computer's inability to quickly judge the contextual information around a news event. Breaking event data incorporates the contextual elements around an event shoring up the deficiencies associated with machine-readable news.
The Technology Advantage
By using sophisticated Semantic web and artificial intelligence-based technology that combines real-time search and entity extraction processes, sophisticated event data providers started using technology to deliver actionable data in an entirely new way. Borrowing pages out of the market data playbook, event-data vendors built systems that delivered new types of information to trading platforms in a simultaneous, low-latency manner, structured to be read by machines.
The evolution of market data is now serving as a model for the development of event data. Trading firms began consuming market data from consolidators, but when automated trading strategies became more prevalent, firms quickly realized the profit potential in consuming market data directly from exchanges. Firms began co-locating their trading infrastructure in close proximity to the venue's market data feeds and order matching engines.
This model can also be applied to event data – placing servers directly at the source of where event data is produced and disseminated allows real-time consumption of the pertinent data around market-moving events and immediate order execution.
In addition to directional event-based trading, firms also use event data to more effectively manage risk. The ability to programmatically monitor important events on thousands of companies is important to the automated market-making segment of the industry that could not feasibly stay on top of breaking events using manual methods. Because it allows investment professionals to filter unimportant extraneous information and digest only critical market-moving data, event data gives greater visibility into events that could impact trading decisions. Hedge funds and asset managers are able to have information about impactful events in real-time, because of its low-latency, structured delivery.
The Future of Data
With high-frequency trading already accounting for more than half of the total U.S. equity trading volume, it's clear that the use of event data will grow rapidly in the years ahead. As this market continues to mature, an even wider segment of the information spectrum – anything from corporate and economic events to geopolitical and softer news stories – could be interpreted for use in traders' electronic strategies.
Though market data will certainly continue to give the best perspective on the state of the securities markets, event data complements firm's real-time perspective into the status of world affairs. It is the natural evolution of the market and is one of the untapped opportunities for asset managers, hedge funds and market makers for increasing alpha and managing risk.
About the Author: Ryan Terpstra is the Founder & CEO of Selerity. He is responsible for the overall strategic direction, operations and leadership of the company. Prior to Selerity, Ryan was the Director of Quantitative News at Thomson Reuters, At Thomson Reuters, he led the design, build and launch of Thomson Quantitative News (TQN), a machine-readable news offering for investment managers. In his spare-time, Ryan enjoys traveling, private aviation and studying classic American & war history. He holds a B.A. in Finance and graduated with honors from Miami University (Ohio) and currently resides in New York City.