Economist's View: Econophysicists

リンク: Economist's View: Econophysicists.

Given recent discussions here, and given that I can't even agree with the claim in the opening sentence that "Predicting financial markets is more of a gamble than traditional economists will admit," I think I'll just present this without further comment:

It's a gamble: UH econophysicists meld science, economics, by Joseph McCauley, Eurekalert: 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, who leads one of the world's most preeminent groups of its kind.

Joseph McCauley, a UH physics professor with a dual appointment as a senior fellow in the economics department at the National University of Ireland, Galway, leads the UH group. The team's main discovery, backed up by empirically based modeling of market dynamics, is that financial markets are unstable. Associate Professor Kevin Bassler, Professor Gemunu Gunaratne and Professor George Reiter – all of the physics department – round out the UH econophysics group that applies their newly discovered models and methods to solve problems in economics.

McCauley will be the only invited physicist to speak in an economic workshop – "Financial fragility and technological progress with heterogeneous agents and social interactions" – Dec. 14-15 in Trento, Italy. He will weigh in with his perspective on the subjects of macroeconomics (the overall aspects and workings of a national economy) and microfoundations (in which the macroeconomic model is built up from the actions of individual agents).

McCauley and his colleagues contend that a market is made up of "noise" in the strictest mathematical sense of a random and persistent disturbance that obscures clarity. Using techniques developed in physics such as entropy – the study of randomness or disorder – challenges the common belief in economics that market statistics have structure and tend toward equilibrium.

"Traditional economics is based far less on empirical studies than its econophysics counterpart," McCauley said. "For instance, deregulation is an example of economists relying on a belief and not hard analyses. Also, histograms – a traditional economist's tool – do not represent normal distributions. Even Nobel Prize-winning economists approach market statistics with a wrong mathematical model already in mind, and the model always fails. Physics helps us understand the information that goes into these models better."

Gunaratne uses the analogy of a pollen grain being heated up in water to illustrate how the randomness of motion is analogous to what happens in a market. Just as a physicist observes the increase and decrease in the temperature of water as variables that agitate or slow the motion of a pollen grain in an experiment, an econophysicist applies these sorts of principles to other such variables in a financial market. In trying to understand this randomness, he said, it is apparent that markets are not bell-shaped curves with symmetry and normal distribution. Instead, financial markets are more like radio static, but with non-bell-shaped noise, with stock prices continually moving up and down in ways that puzzle standard statisticians.

Focusing primarily on the foreign exchange (FX), a 24-hour-a-day traded world market, McCauley, Gunaratne and Bassler say studying the FX yields better information as the largest, most liquid market that dominates other markets because of its sheer volume and volatility. They model both the market dynamics and option pricing by deducing correct models from real market statistics, which is the opposite of what economists do. Broadening the UH econophysics program, their colleague Reiter focuses more on models of the economy, including production and consumption with results that show how individuals' preferences adapt to economic circumstances, a part of reality he said is missing from standard economic models.

"Whatever the specific focus, this relatively young subfield that merges the two disciplines of physics and economics helps us move toward applicable models for use in analyzing markets and economies more effectively and accurately," Bassler said.

Having established one of only a handful of Ph.D. programs across the globe with a specialization in econophysics, UH's physics department in the College of Natural Sciences and Mathematics has recognized a need for educating physics students in this area since the modeling, analytical and computational skills of physicists are exactly the skills needed to study financial markets and the dynamics of the economy in a practical way.

"We realize this is still met with skepticism in more traditional arenas, but we're convinced econophysics will play the leading role as world economies become increasingly more complex and harder to decipher, and the misleading notion of ‘self-regulating markets' will be slaughtered," McCauley said. "To the extent possible in the social realm, we want to create economic theory as science and avoid the ‘mathematized ideology' – as I've coined it – of mainstream economics that is currently the ideology used in the unregulated free market."

Okay, one comment besides the fact that economics is not the stock market. If the econophysicists have something better to offer, great. I'm listening. But until they do have something better, a little less criticism of the models we use would be appreciated (e.g. "Nobel Prize-winning economists approach market statistics with a wrong mathematical model already in mind, and the model always fails. Physics helps us understand..."). Because when your "main discovery, backed up by empirically based modeling of market dynamics, is that financial markets are unstable," I'm not sure that a condescending attitude has yet been earned.

Posted by Mark Thoma on December 2, 2006 at 03:33 AM in Economics, Methodology, Science | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/t/trackback/7012778

Listed below are links to weblogs that reference Econophysicists:

Comments

Well, could you give a refutation of their argument?

Posted by: evagrius | Dec 1, 2006 8:26:32 PM

The physicist Ernest Rutherford is quoted as saying “All science is either physics or stamp collecting”. I see the attitude hasn't changed much in nearly 100 years.

Posted by: Dave Schuler | Dec 1, 2006 8:39:06 PM

"I believe the task of the economist is far more difficult that that of a physicist. A physicist can build upon verified observations and testing to improve knowledge. An economist must deal with human behavior and institutions that are constantly changing and where the rules of interaction are, at best, chaotic. What appears reasonable and verifiable based on a small slice of human interaction considered to be economic in nature may well be just a "special case" situation that can be explained any number of ways... not just by the dynamics of trading or consuming."

http://hallofrecord.blogspot.com/2006/11/no-respect.html

Posted by: Bruce Hall | Dec 1, 2006 8:57:29 PM

Physicists (in general) are a bit arrogant, and this is quite understandable. They've developed theories, mathematics, and predictions that validate todays understanding of some of the universe' most fundamental mysteries.

Witness the crossover success of Luis Alvarez at explaining a cosmic impact to the mystery of the dinosaur extinction - something that had paleontologists mystified for a century or more. Needless to say there was much gnashing of teeth in the Paleo community about this outsider stepping on their toes... OK class, what killed off 90% of all life on earth when the dinosaurs went extinct???

I do think that physicists are too new to the game of attempting to understanding human behavior to be so confident at this point. On the other hand, if a Nobel Laureate class physicist puts his mind to the task, I would expect theories, insights and verifiable predictions on economic hypotheses.

The burden of proof (of theorems) will lie with the physics community of course, but neither should the economic community summarily dismiss them, as the Paleo community did with Alvarez

Posted by: Idaho_Spud | Dec 1, 2006 9:09:49 PM

Remember this information is being processed through a journalist, so the signal to noise ratio is degraded.

The comment "Nobel Prize-winning economists approach market statistics with a wrong mathematical model already in mind..." is, I believe, referring to the fact that Black-Scholes uses a log-normal distribution to model fluctuations, which is known to be wrong empirically. Of course, financial economists also know it's wrong (although probably practitioners were first to figure this out, and the Nobelists Merton and Scholes lost their shirts with LTCM, which tended to assume log-normality in their strategies) - but it makes a tractable first approximation.

Yes, physicists are arrogant and often stick their noses into other fields. But sometimes they make important contributions. Fischer Black was trained in physics, as was Francis Crick, as was Luis Alvarez (already mentioned), as were many Turing Award winners in computer science.

The jury is still out on econophysics. At minimum, it adds to the number of smart people studying finance, though.

Posted by: steve | Dec 1, 2006 9:31:56 PM

Self-regulating markets? Free, unrestricted markets? A market place in which the multitude of choices always results in the greatest good for the greatest number?

You really believe this, don't you?

First, the idea of a free market place is a fiction. Second, even if it could possibly exist, it would result in chaos. No regulations? Gimme a break.

Personally, the econophysics have my vote to give new models a try. The present ones, c/o Friedman, are too simplistic and immature.

Posted by: Stormy | Dec 1, 2006 9:55:30 PM

The confusion seems to be intentional.

It's not that complicated if you focus on the unnatural money creation system.

The Fed inverted the yield curve because that is the power they hold.
-about August.

An inverted yield curve shuts down lending.
-lending has slowed from from above 7% growth to a little above 0%.

A reduction in lending leads to loss of aggregate demand and job losses.
-the jobs part we haven't seen yet, but may soon start.

The unknowns arise from how those in power in other currency zones respond. Will they allow the dollar to depreciate sufficiently to allow the U.S. to find aggregate demand abroad now that U.S. domestic aggregate demand is being extinguished?

Failing to find sufficient foreign aggregate demand will Bernanke continue to wield his power to drive the U.S. into recession/depression with an inverted yield curve or will he drastically cut rates even if core inflation remains untamed?

Will Bush find an excuse to invade Iran or find some other fiscal stimulus sufficient to overcome opposition in congress?

Those that hold power, their future reactions are unknown yet their power is obvious to those that bother to pay attention.

Watch what happens when Paulson and Bernanke go to China. These are not random events, these guys are trying to get others to create economic events favorable for the U.S. so they don't have to.

Posted by: Winslow R. | Dec 1, 2006 10:58:07 PM

Physicists arrogant? There is pretty much a consensus among climate scientists that pumping carbon dioxide into the atmosphere will cause it to warm. Their models are based on physics and chemistry. Following their a-priori, empiricist lite models, different schools of economists could argue forever over what effects a limit on CO2 emissions or acarbon tax will have on the economy. The fact that they don't agree with their colleagues doesn't seem to stop them from sticking the aggregate curve of their noses into this or any other societal debate. In the meantime the planet warms. Friedman, Hayek, Karl Marx -- none were shy about telling us how it is, or ought to be, even if it isn't, or shouldn't. Racial discrimation can't exist! Why? Friedman says it isn't possible, economically. Regulation is bad! Why? Because Hayek concludes, economically, that governments are evil. Freakonomics seems to be the latest effort of the great sages to tell us how it all works. Forget what the psychologists and others who conduct the double blind studies have to say about the matter. Really, apart from Edward Teller, I can't think of a single physicist who is as eager to inject him/herself into important societal debates, as is the average economist. I don't have anything against economics. It's a legitimate and fascinating field of study. I just wish the economists would follow the lead of the physicists and chemists and stay in the lab until they actually prove something.

Posted by: econoskeptic | Dec 2, 2006 12:15:16 AM

I agree it is about the market not about economics. Also it should be by now common knowledge that markets move around due to greed and fear. The traders themselves may be 25 year old's, high on cocaine, motivated more by hefty commissions than risk. The violent swings are often caused by huge amounts of money being poured in to force a particular outcome.

Hence I think a model based on human psychology is a bit more relevant than energized pollen grains. This is another classic case of starting from a bad analogy and trying to prove it.

Posted by: JamesG | Dec 2, 2006 2:16:06 AM

Maybe it's worthwhile to see what Keynes wrote about Planck's remark on economics, which one commentator quoted in "Neoclassical Theory under Fire from the Sciences" entry.

Professor Planck, of Berlin, the famous originator of the Quantum Theory, once remarked to me that in early life he had thought of studying economics, but had found it too difficult! Professor Planck could easily master the whole corpus of mathematical economics in a few days. He did not mean that! But the amalgam of logic and intuition and the wide knowledge of facts, most of which are not precise, which is required for economic interpretation in its highest form is, quite truly, overwhelmingly difficult for those whose gift mainly consists in the power to imagine and pursue to their furthest points the implications and prior conditions of comparatively simple facts which are known with a high degree of precision.(Keynes, Essays in Biography 1951 158n)

BTW, "Complexity: The Emerging Science at the Edge of Order and Chaos", by Mitchell M. Waldrop, may also worth referring. The meeting between top economists(including Larry Summers!) and physicists depicted in this book is quite interesting. But what did Santa fe Institute accomplished in economics since then, beside making Prof. Krugman angry?

Posted by: himagine | Dec 2, 2006 2:31:13 AM

in the fundamental physics department, physicists have been talking about string theory for 35 years or more, today it is the "mainstream", and yet thousands of person-years of work have not produced any testable new predictions. a theory that isn't falsifiable isn't a theory. maybe the physicists have a little work to do at home.

Posted by: supersaurus | Dec 2, 2006 3:56:04 AM

As a former Physicist I would explain this as marketing. Physics Ph.D. provides excellent analytical training, but few jobs. So many physicists have sought and found work in the fiancial industry. University of Houston, which is a second tier Physics school, sees this as a way of attracting physics graduate students. With shortage of jobs for physicists, there is a shortage of graduate students, especially for a second tier school. You should judge this report as you would judge any marketing press release.

Posted by: Roger | Dec 2, 2006 5:49:18 AM

I suspect that Roger is right.

There are some very good physicists working in
this area. Bouchaud is one of the best.

See for example,

http://arxiv.org/PS_cache/cond-mat/pdf/0504/0504079.pdf

(Mark, I'd be interested in what you think of the
above paper.)

For more of his papers, search bouchaud on

http://arxiv.org/find/cond-mat

Posted by: Jeffrey Miller | Dec 2, 2006 6:30:50 AM

Here are some more papers. To avoid
having to resgister, click on the "e-print"
link below the abstracts to get the papers:

http://www.cfm.fr/us/publications.php?subject=2

Bouchaud and his colleagues also run a sucessful hedge fund.

Posted by: Jeffrey Miller | Dec 2, 2006 6:35:03 AM

Roger leads me astray.

I'm waiting for the blackhole analysis where it is explained by former cross-over econophysicists unable to find gainful employment in either Physics or Economics that one's ubiquitous housekeeping chores can be accomplished by judicious application of anti-entropic analysis (AEAE)[pronounced like the primal scream].

So Economics accepts physicists. Prolly mathematicians. A few chemists and the odd biologist. What about historians?