Web data predict stock movements: online reading behavior can help warn of financial crises.
Web surfing patterns of people who are reading financial news can be used to make accurate predictions of stock movements up to a couple hours in advance, a new study suggests. The technique may one day help financial authorities monitor markets and fend off emerging crises.
A team of physicists led by Gabriele Raneo of the IMT Institute for Advanced Studies in Lucca, Italy, suspected that better stock predictions might be made by looking at more than the positive or negative sentiment expressed in an article about a company. Another key indicator might be how many people actually click on links to an article, a sign of the article's social influence and how much readers are paying attention.
The team used a year of data collected in 2012-2013 from Yahoo! Finance, an online portal for financial news and data. Looking at the 100 U.S. companies with the most frequent mentions in news articles, the team calculated a measure of sentiment for each article. The team then weighted that measure to reflect reader behavior: Articles counted for more if more readers clicked on links to them.
The result was a moment-by-moment signal for each company showing how sentiment and interest fluctuated during the day, and how strongly. The team compared these signals with actual market fluctuations of prices, volume and volatility for the stocks of the 100 companies. The signals offer a significantly improved predictive capacity for stock movements, especially for movements roughly one hour later, the researchers conclude January 25 in PLOS ONE.
Mathematical finance expert Rosario Mantegna of the University of Palermo in Italy suggests that this method could prove useful for financial authorities worried about the potential for explosive financial events--a bank run triggered by a surge of investor fear, for example.
An important aspect of the work is its ability to monitor web activity on timescales as short as a minute, says study coauthor Guido Caldarelli, also at IMT. This kind of analysis, he says, could be carried out in real time if authorities had access to the data.
More generally, the study illustrates the potential of "big data" to provide new means to detect broad social patterns of belief in areas ranging from public health to political polling. "Questionnaires are slow and people don't always give their real views," he says. "An advantage of web data is that people tend to be more sincere when they're browsing."
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||MATH & TECHNOLOGY|
|Date:||Feb 20, 2016|
|Previous Article:||Altered milk protein cleans up pollution: amyloid fibers can snatch heavy metal contaminants from water.|
|Next Article:||Rapid spread of Zika virus raises alarm: disease linked to birth defect is pushing northward from Brazil.|