"Veiled Agendas" of News Stories -- and Oil Traders
With uncanny timing, a novel released just before the BP spill looks at how tangential data and language influence the markets.
It might seem strange for people to ask a newly minted novelist what he thinks about the BP oil spill's effects on oil futures, but it makes more sense when your novel is called Kapitoil, as mine is, and the plot revolves around a titular computer program that sifts through and analyzes news articles -- mostly about political instability in the Middle East -- to forecast, in roundabout fashion, oil futures.
Now, sadly, we're confronted with a disastrous nonfiction news event that cuts out the middleman and has had a direct impact on spiraling oil prices.
As analysts have pointed out, oil futures have plunged for a few reasons: the sinking Euro; the negligible impact the spill has had on oil supply (in fact, US oil production has actually risen since the accident); and the fear that future offshore drilling projects will now take longer to gain regulatory approval.
Upon hearing about the accident, insiders and laymen alike probably could have assumed that a devastating accident provoking global headlines would make speculators less sanguine about an industry's prospects.
The fictional program in Kapitoil, however, ignores this "front-page news," as its protagonist, Karim Issar, a young Qatari who has learned English through the worlds of finance and technology, calls it. If everyone else is employing the same information, he reasons, then his program will have no advantage.
Rather, Kapitoil's algorithms focus on the less obvious, back-page stories -- the "tangential data" -- and dissect the diction of these articles to find instances of subjectivity that might subliminally influence readers. An example he uses: "The Prime Minister, after a round of bitter questioning, appeared weary and resigned." To Karim, the words "bitter," "weary," and "resigned" are as valuable -- if not more so -- for oil forecasting than words everyone else fixates on, such as "spill," "accident," and "environment." The program then weights the power of individual words against historical data and predicts how their usage in recent articles will alter the futures market.
Karim hypothesizes that traders read these articles and unconsciously react according to their veiled agendas. Once they do so, others follow suit, whether they have read the same articles or not. With a little literary license, it's not so far off from high-frequency trading, the tactic of having computers make trades based on electronic data before humans have a chance to receive it, which some critics have pinpointed as the source of the May 6 crash.
Is a program like Kapitoil really feasible? Can stock market algorithms truly break down language with this degree of sophistication? Textual-analysis programs have existed for almost two decades, though they have almost exclusively focused on financial news. Overnight Desk, LLC, performs a Kapitoil-like service nightly, curating and synthesizing the most influential financial -- not strictly political -- articles from around the world and sending reports to its clients. As of now, founder Brad Stoler is confident that humans can do a better job than computers at culling and analyzing relevant material.
Yet two professors, Robert P. Schumaker and Hsinchun Chen, have developed a new system, Arizona Financial Text (AZFinText), which parses language in financial articles in a manner quite similar to Kapitoil. In tests on 2005 stock market data, it outperformed every mutual fund that traded in the same securities and was the fifth-highest-performing quantitative mutual fund in the world. In another parallel to Kapitoil, their verbal classifications come from a Defense Advanced Research Projects Agency news-analysis project centering on national security -- which incorporated news about Latin American terrorism.
Ultimately, though, Kapitoil isn't about the viability of a terror-centric linguistic-analysis program. It's a metaphor for how the news can't possibly achieve objectivity, no matter how much it pretends to; how language is inevitably slanted; and how we, its consumers, can't help but absorb this bias. (It's also about the difficulty of human connection, the textual analysis of Bob Dylan lyrics, and fantasy baseball.) The novel is set in 1999, but its implications target the post-9/11 era, when the rapacious 24-hour news cycle has become as newsworthy as what it covers and plays an increasingly significant role in steering voters -- due to both Fox News/MSNBC-style heavy-handedness and the more pernicious, seemingly moderate outlets that plant their biases with greater subtlety.
So, will this article about oil futures have any effect on oil futures? No; it's far too detached from any real economic reporting. But if Kapitoil actually existed, that very detachment, ironically, might just make it tangential enough to be centrally important.
Copyright © Teddy Wayne 2010
The information on this website solely reflects the analysis of or opinion about the performance of securities and financial markets by the writers whose articles appear on the site. The views expressed by the writers are not necessarily the views of Minyanville Media, Inc. or members of its management. Nothing contained on the website is intended to constitute a recommendation or advice addressed to an individual investor or category of investors to purchase, sell or hold any security, or to take any action with respect to the prospective movement of the securities markets or to solicit the purchase or sale of any security. Any investment decisions must be made by the reader either individually or in consultation with his or her investment professional. Minyanville writers and staff may trade or hold positions in securities that are discussed in articles appearing on the website. Writers of articles are required to disclose whether they have a position in any stock or fund discussed in an article, but are not permitted to disclose the size or direction of the position. Nothing on this website is intended to solicit business of any kind for a writer's business or fund. Minyanville management and staff as well as contributing writers will not respond to emails or other communications requesting investment advice.
Copyright 2011 Minyanville Media, Inc. All Rights Reserved.
Daily Recap Newsletter