A London-based hedge fund is developing a trading model that will track
sentiment on Twitter to predict daily fluctuations in the stock market,
Derwent Capital Markets will base the model on research
published by the University of Manchester and Indiana University in
October, which found that the rise and fall of "calm" words expressed
on Twitter are followed, 88 percent of the time, by corresponding
shifts two to six
days later in the Dow Jones Industrial Average's closing price.
In an interview with Bloomberg, the hedge fund's
co-owner, Paul Hawtin, explained, "Sentiment and mood dramatically
change the impact of positive and negative news stories. If the
market's in a very positive and bullish mood, it can shrug off bad
news--bad news comes out and you expect the Dow to fall, and it
Is the hedge fund on to something?
- I'm Intrigued But Skeptical, states
Felix Salmon at Reuters. It makes sense conceptually that stocks fall
when people are nervous and selling and rise when people are calm and
betting on the future, he concedes, and Twitter does excel at "giving a
real-time glimpse into the sentiment of millions of people." But
Predictive power isn't enough to make money in the stock market: it's perfectly possible to make money 87.6% of the time but still lose money over the long run. I also suspect that these kind of algorithms are going to have difficulty keeping up with Twitter as it evolves away from people broadcasting the minutiae of their lives, and towards more sophisticated conversations.
- So Am I, agrees
Barry Ritholtz at The Big Picture. He notes that the research
underpinning the trading model was conducted between
February and December 2008, a short and atypical period in which the market collapsed:
The monkeys were unusually agitated during this time. This model will need to prove itself during periods of normal volatility and sentiment. Still, it would not surprise me if the Twit stream reveals some consensus about the mood of the monkeys. At extremes, that could be an assist in developing a trading thesis about intermediate term highs and lows ... I wonder if it can be used to manage day-to-day trades.
- The Model Can Be Manipulated, points out
Mike Taylor at The New York Observer. Someone could register "a whole
bunch of Twitter bots designed to tweet certain words and make the
Derwent machine go insane," he explains.
- I'm Not Sure What Joke to Make, says Pascal-Emmanuel Gobry at Business Insider: "Finally a business model for Twitter? A sure sign of a bubble? But where? On Twitter? Or on the stock market? Or both?" Gobry also notes that the story doesn't mention StockTwits, a platform that tracks stock discussions on Twitter. "It would be ironic if StockTwits was less good than Twitter at predicting stock movements," he adds.
- Derwent's Quest Is Not New, argues Matthew Ingram at GigaOM:
Stock traders have always looked at market-sentiment indicators, including some who study fashion trends or popular music to try to determine which way stock indexes are likely to go. It’s a little like science and a little like voodoo, as even some traders will admit. But the key is information--as much information as possible--and investment banks and hedge funds now have more of it than they could ever want, and the computing power to take advantage of it by crunching and analyzing it.