Turning Up the Heat
Leading Wall Street firms with algorithmic or quantitative trading groups in equities already incorporate news feeds into their trading applications. One news tracking, filtering and aggregation service catching the attention of major sell-side firms is Relegence, which scans over 18,000 business-centric sources, including domestic and foreign language newspapers, business media, national news sites, stock exchanges, government sites such as the SEC and FDA, corporate Web sites and blogs.
Relegence calculates a news heat index, called Relegence FirstTrack, that counts the number of news stories on a given company and quantifies on a scale of minus-four to plus-four when there is "heat" on a given stock. According to Relegence (New York) founder and CTO Edo Segal, "There is a way to automate detection of abnormal behavior in the news domain that is indicative of volatility and really is a very important signal in any quantitative approach to algorithmic trading."
Bear Stearns' E.A.S.T. division -- which covers algorithms, market-impact models and alpha models, and anything quantitative that relates to equity trading -- distributes the Relegence feed internally to its trading applications on its block and portfolio trading desks. "We're using [heat] as an indication that perhaps there's something going on with that security that the traders should look at and in combination with seven other factors that the trader needs to look at, like volume, momentum or the overbought/oversold indicator," explains Matthew Celebuski, senior managing director in Bear Stearns' E.A.S.T. division. In addition, institutional clients also can query the heat data through Bear Stearns' systems.
Bear Stearns digitally consumes the heat indicator information for all domestic and international securities, adds Celebuski, whose team has built tools on top of the heat index that are customized for both block and portfolio traders. "Suppose someone on CNBC mentions the company IBM and 'merger' in the same 30-second period -- Relegence would pick that up," he says.
Though Bear Stearns has not yet built a full-blown quantitative model that incorporates the news signal, the firm's Cherasia says it has incorporated the heat statistic into its pre-trade analysis system, which makes a recommendation on which strategy would be the best to trade a particular stock at a particular time. "In fact, about a third of the time, the recommendation is not to use an algorithm," notes Celebuski.
Ivy is Editor-at-Large for Advanced Trading and Wall Street & Technology. Ivy is responsible for writing in-depth feature articles, daily blogs and news articles with a focus on automated trading in the capital markets. As an industry expert, Ivy has reported on a myriad ... View Full Bio