Our friend John of http://www.aprendiendolabolsa.com.ar/ sent us this note:
The Man with the Bag
The econophysics is a new discipline, using mathematical models, fractals and chaos theory, discusses the not-so-random fluctuations in financial markets.
By Federico Kuks
There is no doubt that the world is complex. Although it is not just for its inherent complications, wars, power struggles and increasing distance between rich abysmal y pobres.
Estrictamente hablando, nos rodean los “sistemas complejos” caracterizados por la autoorganización de sus componentes y por la aparición de rasgos o características emergentes. Los ecosistemas, el cerebro, la atmósfera, las corrientes oceánicas, los hormigueros son sistemas complejos frente a los cuales la mejor forma de aproximación es la holística.
La economía es otro ejemplo clásico, capaz de ser abordado a partir del análisis de su comportamiento fluctuante, sinuoso e impredecible del todo. La psicología, la estadística y obviamente la teoría económica intentan a diario entender la turbulencia de los mercados and avoid certain financial guidelines. In this tandem of science was added not long ago a new one, the "econophysics", who leads his theoretical arsenal (fractals, algorithms and chaos theory) to embrace the economic complexity.
"I'm always interested in any social application of science. Trying to understand other things than those microscopic things that are difficult to approach and touch. And social issues are indeed complex and there are many problems to do modeling. You can contribute a lot by using the tools of physics to tackle problems that are of interest to economists, "says physicist Martin Zimmermann.
- How long is physical?
-I am a doctor in physics for 10 years. I got here but I doctor's degree in Uppsala, Sweden. And then I majored in Spain.
He majored in ...
"In what is called" dynamic systems ", that is, complex systems that any change or development of the state at a time. You can define a set of elements that interact. With a English colleague, Victor Eguíluz, we were interested the issue of financial markets and what is known as the "problem of cooperation", ie why in a non-partisan group of individuals may come and kinship cooperation exist where, in principle one would expect the optimal action is to not cooperate. This is the famous prisoner's dilemma.
"That's game theory.
"Exactly. We start from the idea that the economy is and functions as a complex system. In complex system actually is a very comprehensive label: a complex system is the water that you take, is the economy itself, an ant, an ecosystem, etc. Complex these systems are called for to analyze and understand you have to look at as a whole, with a holistic view. If you attack a problem and you can not break down into their simplest elements, there you may face a problem of complex systems. We are immersed and surrounded by these systems.
- what did they do?
-work with a very simple mathematical model, or as we like to say to the physicists, a "toy model" to mean a model that describes very few items and has a behavior that resembles that observed in financial markets. More specifically, the phenomenon that was trying to understand why fluctuations or changes in the prices of stocks and bonds in the stock have a type of behavior is not purely random. In recent years, many physicists measured this and found that the fluctuations are larger than what comes in the models of economists. There are large fluctuations unlikely. But if you really what you measure in the large fluctuations are relatively quite likely.
"But economists usually deal from their experience.
"Mmm, not always. Our goal was to provide a possible explanation why these fluctuations are sometimes larger than expected. Model showed that the actions of the agents (brokers) occurred in a herd or flock of sheep. If one took the action of an agent, the whole that this agent was connected performing the same action. Because of that herd behavior, these large fluctuations occurred.
- That would predict how financial markets fluctuate?
"No, unfortunately. It would be interesting if you predict will raise or lower the value of the shares. The model was really aimed at measuring variations. The we find is that the stock market behaves in a manner similar to a pack. Was a result mostly schematic.
"So, it worked with a sample.
-not trying to accurately represent reality, but that "the shape of the distribution, changes and fluctuations looks like this." And that's what we did. Currently there are models that fit more.
-also these models are valuable tools and representations.
"Yes." It's not something provable because I analyzed a real series. Yes worked after the Santa Fe Institute in New Mexico (United States) with all the data that you can have on the London Stock Exchange. And it was amazing because you could see the behavior of traders seriously. It was interesting to see what they were doing and why they were doing.
"But was not that.
-No. I set out to design models that allow predictions of financial time series. And now with a colleague are working on it as a venture to provide a consultancy style service, a financial service. I could say "my model predicts that in the coming weeks, the dollar goes up or go down. "There are many people who have interest in high-risk investments and that would be a good help for that.
- How do physicists to economists?
-worlds are apart. A physicists who have studied a little economics surprise us some of the assumptions that economists use. Sometimes criticize little their principles, their hypotheses. What we see is they have a very strong willingness to work I have a analytical resolution in closed form. The theoretical economists point to prove theorems, which in complex systems is something recontra difficult.
"And that there are frictions.
"Sometimes. Another of my specialties are computer simulations. Since I have devoted my program and use it for research. And I work a few years ago to develop software for optimization, applied to logistics. Along with the company Cybermapa we are marketing a product. Our endeavor is called Zews ( www.zews.com.ar ).
Returning to the crossings between physics and economics, is a very green area, not exploited, right?
"Yes." A physicists quite difficult for us to publish in journals of economics because they expect to appear theorems, a demonstration.
- And what kind of people point with their consultant?
-A people who have some knowledge in financial assets. I sell "signals."
"As an oracle.
-autocatalogo I really like "technology entrepreneur." What really I do is run a few programs on the computer to tell me about how the dollar will fluctuate. The advantage of such procedures is that we abstract of emotion and we run the algorithm. That is good because if you have experience in trading, buying and selling, you realize that emotions play a very important time to say "what?"
"So, the market is not a field 100 percent rational.
"No, there is an irrational component to explain exactly why they form the waterfalls or flocks. Is what is called the "risk averse" when it starts to be some sort of sell-off, arm a larger wave, and there is a multiplier effect easy to notice. When there is some level of inflation, as in 80, there is a very difficult behavior to stop, it's like a snowball.
"For what is seen physicists are welcome in various fields.
"Yes, for example I'm also working with biologists, agronomists and ethologists by network models of insect pollinators. Is a complex system because you have many insects and plants interact. Are mutualistic networks because the insect gets the benefit of a plant and pollinate the plant receives the benefit of pollination. The field samples showed a pattern of very specific interactions.
- "The idea is to build a model that seeks to represent that phenomenon?
"Exactly. It is interesting to work with people from other specialties as they have a different view of doing science.
-would also be a mutual network.
try to be mutual, but sometimes these links TURNS competitive relationships.
source:
Salu2.
Paul Devaux.
0 comments:
Post a Comment