As unrest sweeps across the Middle East and oil prices climb, today’s economic climate could make any investor nervous.
But not Joey Engelberg.
Engelberg, an associate professor of finance at UNC, has conducted a research project with two of his former classmates at Northwestern University to gauge how well Google search trends can predict hot stocks.
The researchers used data from the website Google Trends to show that frequently searched terms can predict investor attention —which in turn causes temporary price spikes in the market.
Engelberg worked with Zhi Da and Pengjie Paul Gao, both professors at the University of Notre Dame, to publish “In Search of Attention,” a study relating Google and stock market trends, in December.
“The Google paper was born of a need for a better measure of investor attention,” Engelberg said.
He said investor attention is traditionally measured based on how often stock names appear in financial journals and on trading volume.
But those measures can be skewed predictors.
“No one reads every single article in the Wall Street Journal on a daily basis,” Da said. He said media focus on a stock doesn’t mean investors take interest in it.
Trading volume is also a poor measure, since single investors can trade large shares of stock for reasons not related to the market.
But Engelberg said Google provides a real-time measure of direct investor attention.
He said that while sophisticated institutions have market informants, ordinary people — called retail investors — do financial research through search engines, making Google data a valuable indicator of their interest.
“Attention shocks lead to temporary spikes in prices, and we found evidence for that,” he said.
Engelberg said the study’s findings have already been used by at least one major hedge fund, and Da said other academics have begun to further explore the data’s financial implications.
But Da said the trio pioneered their field of research.
“Joey, Paul and I are among the first to apply Google search to financial data,” Da said.
Engelberg said he became interested in the impact search engine trends have after reading an article in the journal “Nature” that used data to predict flu outbreaks.
“I thought it was a brilliant idea,” he said.
Engelberg, Da and Gao attended graduate school at Northwestern University together and have worked together in the past. It was their common interest in predicting investor attention that prompted them to research together.
Now, they are working on other projects that use Google data to predict financial outcomes. One measures a “fear index” by noting how often people search economy-related terms like “recession” to create an index of investor sentiment.
“If you use this data in the proper context, it can unveil a lot of information we never used before,” Gao said.
To get the day's news and headlines in your inbox each morning, sign up for our email newsletters.