Predicting stock markets with words

Nieuws | de redactie
29 juli 2011 | To taxpayers, financial markets are unpredictable money-eaters. Financial reporters claim to know where things are moving. Now researchers show it does not matter so much what they say, but how often they say it.

Forecasting the future can make a poor man rich on Wall Street.What instruments you rely on to estimate where stock markets willgo depends on your taste and expertise.  Two Irish researchersnow added another tool. It measures how homogenous the language isthat financial analysts and journalists use when writing aboutfinancial markets.

Aaron Gerow (Trinity College, Dublin) and Mark Keane (UniversityCollege Dublin) examined 18.000 online articles from theFinancial Times, The New York Times and BBC between 2006 and 2010.Their results showed that the closer stock markets get to theirhigh point, the more similar becomes the language used fromfinancial reporters.

Expressions like ‘stocks rose again’, ‘scaled new heights’ and’soared’ are then most frequently used. In another observation,Keane remarks that “[financial reports] also appear to refer to asmaller-than-usual set of market events – presumably because of anincreased fixation on a small number of rapidly rising stocks.”Reporting on ever-rising stocks from Google and Apple becomes veryfashionable accordingly.

Directly after the market bubble busts, this trend is reversedand as the stock market falls, journalists’ language becomes morediverse. Why this would occur is not completely clear to the Irishresearchers. “Maybe it’s a bit like happy families are all happy inthe same way, but unhappy families are unhappy in many differentways,” says Keane referring to a quote from the Russian writerTolstoy.

Plotting their frequency indicator together with the Dow JonesIndex becomes more insightful. The red line depicts the frequencydistribution of words used by financial analysts and reporters. Thelower the red line, the less similar the language used. The stockprice index, meanwhile, is represented by the blue line.

Stock Market Graph

Source: Aaron Gerow and Mark Keane, “Mining the Web for the’Voice of the Herd’ to Track Stock Market Bubbles”

Schrijf je in voor onze nieuwsbrief
ScienceGuide is bij wet verplicht je toestemming te vragen voor het gebruik van cookies.
Lees hier over ons cookiebeleid en klik op OK om akkoord te gaan