As Economy Tanks, Calls For More Big Data

Some practitioners of the “dismal science” are dismayed by what they contend is flawed, outdated and incomplete data used by the U.S. Federal Reserve to make its economic forecasts. As global stock markets gyrate in response to the economic slowdown in China, these economists argue that the Fed should be making greater use of readily available real-time economic data.

Among these sources is a Massachusetts Institute of Technology initiative dubbed the “Billion Prices Project” that measures inflation using real-time data gleaned from online purchases tracked on retailers’ web sites. A Google Price Index and other search engine tools offer similar retail detail, economists note.

Meanwhile, real-time data proponents note that researchers are mining social media sites like Twitter looking for other leading economic indicators. A particularly useful Twitter hashtag, #NFPGuesses, for example, contains a weekly aggregation of estimates on non-farm payroll gains.

Elsewhere, the online real estate service Zillow tracks home sales and mortgage lending, providing real-time data on the critical and volatile housing and mortgage sectors.

Job data is said to be readily available from Internet job searches and job postings. Economist Tara Sinclair argued these data “could be used to predict employment for the following month.”

Added Sinclair in a recent column published by The Japan Times: “Properly used, new data sources have the potential to revolutionize economic forecasts. In the past, predictions have had to extrapolate from a few unreliable data points. In the age of big data, the challenge will lie in carefully filtering and analyzing large amounts of information. It will not be enough simply to gather data; in order to yield meaningful predictions, the data must be placed in an analytical framework.”

The mission of another real-time data tracker called PriceStats is “to transform data into meaningful tools for economic analysis.” Among the data techniques used by the site to develop data-driven economic indicators is web scraping to monitor daily retail price fluctuations. The technique involves converting unstructured data, usually in HTML format, into structured datasets that can be stored and analyzed.

It also tracks key online retailers to measure inflation and “purchasing power parity ” trends. Once inflation and PPP data are processed, the tracker compiles daily economic statistics using econometric techniques.

Despite ample evidence of real estate and commodities bubbles in China, most economists failed to spot the current market troubles that have sent China’s stock market into free-fall. Nevertheless, a growing number of economists argue that big data techniques and greater use of available real-time economic statistics must be leveraged so policy makers can better anticipate downturns.

“In order to improve economic predictions, economists must be encouraged to seek new sources of data and develop new forecasting models,” added Sinclair, who also teaches economics at George Washington University. “As we learn how to harness the power of big data, our chances of predicting — and perhaps even preventing — the next recession will improve.”

Recent items:

Big Data’s Evolving Role in Government Economic Policy

The Intersection of Public Policy and Big Data