No crystal ball necessary: New tool IDs predictable economic variables

July 21, 2009,

You don't need a crystal ball to tell you what is going to happen next in the economy. You need a statistical model. A new method from North Carolina State University can help researchers determine which economic variables they should focus on by identifying whether a variable can be predicted.

The new method could be a breakthrough for economists and statisticians, says Dr. Mehmet Caner, an associate professor of economics at NC State and co-author of the paper unveiling the new research. "This could lead to much great insight into the national and global economy by providing the impetus to develop new forecasting models for those variables that can be predicted," he says.

Currently, economists and statisticians use tools called "unit root tests" to determine whether an economic variable - such as unemployment - can be predicted. "The issue," Caner says, "is that unit root tests often say variables are unpredictable when they actually can be predicted.

"I think our method will show that many variables believed to be unpredictable are actually predictable," Caner says, "including currency exchange rates and gross domestic product."

Caner worked with Dr. Keith Knight, of the University of Toronto, to develop an entirely new method using new models to better differentiate between predictable and unpredictable economic variables.

The idea stemmed in part from the use of Bridge estimator models in medical research to identify sections of genetic code that may be related to disease. Caner and Knight saw that similar techniques could be used to determine which economic variables within an economic system might be predictable.

Caner will present the new research, "No Country for Old Unit Root Tests: Bridge Estimators Differentiate between Nonstationary versus Stationary Models and Select Optimal Lag," at the European Economic Association and Econometric Society European meeting being held in Barcelona, Spain, from Aug. 23-27.

Source: North Carolina State University (news : web)

Explore further: New model predicts hot spots for mercury in fish

Related Stories

New model predicts hot spots for mercury in fish

December 1, 2008

Mercury levels in fish are prompting widespread consumption advisories and uncertainty among consumers over which species are safe to eat. Now researchers at North Carolina State University have developed a model that will ...

It's a gamble: Econophysicists meld science, economics

November 30, 2006

Predicting financial markets is more of a gamble than traditional economists will admit, and making sense of such numbers is more like trying to decipher noise blasting from a loudspeaker, says one University of Houston econophysicist, ...

Why conservation efforts often fail

September 18, 2007

Modern conservation techniques have brought us the resurgence of American bald eagles, sustainable forest harvests and the rescue of prized lobster fisheries. So how can modern conservation strategies also have wrought such ...

How Will the Economy Fare in 2007?

January 17, 2007

North Carolina State professor Dr. Michael Walden is an economist and not an airplane pilot, but he uses aviation terminology to discuss the prospects for the U.S. economy in the next 12 months.

Recommended for you

Powering a pacemaker with a patient's heartbeat

February 20, 2019

Implantable pacemakers have without doubt altered modern medicine, saving countless lives by regulating heart rhythm. But they have one serious shortcoming: Their batteries last only five to 12 years, at which point they ...

The global impact of coal power

February 20, 2019

Coal-fired power plants produce more than just carbon dioxide, which contributes to global warming. Coal burning also releases particulate matter, sulphur dioxide, nitrogen oxide and mercury—thus damaging the health of ...


Adjust slider to filter visible comments by rank

Display comments: newest first

not rated yet Jul 21, 2009
Well, at least they should have plenty of data to test their theories on. It should be interesting if they can model the current economic crisis. Maybe once and for all the Democrat vs Repub philosophies can be tested.
5 / 5 (1) Jul 21, 2009
The primary problem is they support their party over their nation they aren't looking to make the situatin better they just aim to help out their campaign contributors

I pledge allegiance to the party and to the chicanery for which it stands, one nation under our thumb with prosperity and justice for none
not rated yet Jul 21, 2009
Economics is one of those things, like AI or climate and/or weather, I don't think we're going to be able to accurately and reliably model them until we have quantum computers. There's too many variables within variables within variables.

It's always bothered me when people refer to economics as a science anyway. There's general ideas and reasonable principles but I don't think there's any immutable laws, like physics. It's still just a freaking giant ongoing experiment, much like the various civilizations.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.