Physicists fail on Wall Street

[Michael]

This one's for Dominck! :D

Well, we've all been reading about how Wall Street made lots of stupid bets on the economy - mostly due to massive failure of the assumptions that inform their mathematical models of risk.

Does anyone realize that it has been a bunch of physicists causing most of the damage? (I've known about this for a long while - these are the guys who came up with the models based on endless growth in US housing prices, and of course the famous 'risk' model that had all the banks, hedge funds and investment houses following the exact same 'risk' strategy (which effectively concentrates the risk factors and ensures they all suffer when something goes wrong).

It is to be noted here that this is not the case of a bunch of financiers taking the mathematical models from the physicists and (mis)applying them to the world of finance. It is the physicists bringing the mathematical models to Wall Street.

They are known as “quants” because they do quantitative finance. Seduced by a vision of mathematical elegance underlying some of the messiest of human activities, they apply skills they once hoped to use to untangle string theory or the nervous system to making money.

This flood seems to be continuing, unabated by the ongoing economic collapse in this country and abroad. Last fall students filled a giant classroom at M.I.T. to overflowing for an evening workshop called “So You Want to Be a Quant.” Some quants analyze the stock market. Others churn out the computer models that analyze otherwise unmeasurable risks and profits of arcane deals, or run their own hedge funds and sift through vast universes of data for the slight disparities that can give them an edge.

Still others have opened an academic front, using complexity theory or artificial intelligence to better understand the behavior of humans in markets. In December the physics Web site arXiv.org, where physicists post their papers, added a section for papers on finance. Submissions on subjects like “the superstatistics of labor productivity” and “stochastic volatility models” have been streaming in.

Article (http://www.nytimes.com/2009/03/10/science/10quant.html?_r=1&hp=&pagewanted=all)

Indeed - this is not a good trend. I predict that over the medium term, banks that eschew such funky math models will produce a more steady and reliable source of profit than those who use them. So far, this has been the case.

I'm a strong opponent of this kind of systemic approach. I think people are more than numbers and I don't like treating poeple as if they are abstract numbers. That's inhuman, besides being problematic for long term profits.

The irony is that these people are trying to bring the principles the physical sciences and applying them to human beings and in the process, producing worst results than the traditional 'art' that they've long criticized for not being methodical or predictable enough.

[Dominick]

This one's for Dominck! :D
Why ? :shrug:

He says exactly what I'm saying :

As Dr. Derman put it in his book “My Life as a Quant: Reflections on Physics and Finance,” “In physics there may one day be a Theory of Everything; in finance and the social sciences, you’re lucky if there is a useable theory of anything.”

Who was making decisions based on blatantly clearly pre-announced probability techniques ? Not the 'quants', but the greedy bankers. There is nothing wrong with these techniques if they're used for what they are : research. It's the idiocy of the economists, bankers and politicians that transposed probability to reality that's the problem.

[Michael]

Why ? :shrug:
You were going on about the silliness of mathematics used in economics.

As it turns out, it is a bunch of physicists who are doing the math that has driven Wall Street into bankruptcy.

In other words, a bunch of physicists doing exactly what you critique in another thread - using complex models and applying them to economics (and producing horrendous pseudo-scientific results).

According to your argument in the other thread, this should have helped to turn economics into a "real" science.

As it is, economics is a human science and will remain so. All the math and stats in the world won't ever change that (despite the desire of so many 'science-types' to change this).

[Dominick]

I was still editing my post when you replied. Maybe you'll need to edit too to reflect the changes.

Anyway, I've just ordered My life as a quant (http://www.amazon.com/My-Life-Quant-Reflections-Physics/dp/0471394203) from Mr. Berman and I strongly suggest you do the same. And no, you don't get to skip the chapters on the mathematical models. :D They're the very essence of this discussion.

[Michael]

I was still editing my post when you replied. Maybe you'll need to edit too to reflect the changes.

Anyway, I've just ordered My life as a quant (http://www.amazon.com/My-Life-Quant-Reflections-Physics/dp/0471394203) from Mr. Berman and I strongly suggest you do the same. And no, you don't get to skip the chapters on the mathematical models. :D They're the very essence of this discussion.
I have no interest in those who seek to reduce the human arts into mathematical equations so they can be manipulated.

The human arts are best when the are approached with humanism.

P.S. The games that people will play in order to try to swindle a buck will never surprise me. That's what these physicists were doing. Trying to get rich with a swindle. Nothing new or surprising there - just the scale of the damage.

[Dominick]

I have no interest in those who seek to reduce the human arts into mathematical equations so they can be manipulated.

The human arts are best when the are approached with humanism.

P.S. The games that people will play in order to try to swindle a buck will never surprise me. That's what these physicists were doing. Trying to get rich with a swindle. Nothing new or surprising there - just the scale of the damage.
That's not at all what happened.

It starts with the commercialization of science, the pseudo-charity grant system and the cease of funding of fundamental research. Most scientific institutions were put under control of managers instead of scientists in the early eighties. The result of this was that they started applying 'economical' concepts to these institutions. For instance, they wanted ROI, return on investment, they wanted immediate and circumscribed goals and results, and so on. For that a grant system was introduced; a scientist gets funding for one year provided (s)he makes those goals.

Now, for applied science that may work, but for fundamental research that is complete bullshit. Fundamental research never knows where it will lead to, and even less how long it will take. People who do not understand or accept that have no business in the field. So that approach was a complete disaster on all accounts and for that reason much of fundamental research was scrapped rather than left alone. An example is the scrapping of the American version of a Large Hadron Collider.

Those -politico-economical- decisions left many researchers out of work. And of all people it was Wall Street that started to hire them. The tools that these former physicists started to develop were the right tools for the job. Infinitely better than that what was used previously (and which has always been the subject of my critique). But these tools were both fundamental research and probabilistic in nature. Fundamental research is synonymous with "Do not try this at home", meaning it's research and nothing more. The decision to actually and already apply this fundamental research to the real world was of course exclusively a managerial decision. Secondly, from probabilistic results, those specific datapoints were extracted and elevated to precise predictions which suited management, i.e. the bankers the best. Again, it's entirely their responsability to misapply, misunderstand and/or abuse technical results.

It's entirely the responsability of management, not of the employees. In other words of the bankers, not of the once-physicists. It's also just one devastating example of the madness of (over-)specialization, the insanity of applying 'economical' concepts to everything, and of the massive incompetence of politicians who started the deterioriation of fundamental research.

Clinton got it wrong. The phrase should not be "It's the economy, stupid" but "It's the stupid economy".

[Michael]

Sorry, your post is just an apologetic for physicists (which is what I expected and why I cited the post as "for Dominick" - I knew you would try to apologize/exonerate them - your bias here is predictable and expected). ;)

I posted the thread in fun - the irony of the topic obviously escapes you.

You objected (in another thread) that economics is using funky math that isn't rigorous enough for you, suggesting that economics needs to adopt the mathematical rigeur of the hard sciences (such as physics). When it is shown that is exactly what they are trying to do and that's what is causing most of the problem in financial services, you switch over to apologetics for physicists, despite the fact that they were (as a group) substantially involved in the process

That's almost as funny as the original article!

Shall we continue this discussion in the Alternative Economic theory thread? This thread is just an illustration of that phenomenum.

[Dominick]

Sorry, your post is just an apologetic for physicists (which is what I expected and why I cited the post as "for Dominick" - I knew you would try to apologize/exonerate them - your bias here is predictable and expected). ;)

I posted the thread in fun - the irony of the topic obviously escapes you.

You objected (in another thread) that economics is using funky math that isn't rigorous enough for you, suggesting that economics needs to adopt the mathematical rigeur of the hard sciences (such as physics). When it is shown that is exactly what they are trying to do and that's what is causing most of the problem in financial services, you switch over to apologetics for physicists, despite the fact that they were (as a group) substantially involved in the process

That's almost as funny as the original article!

Shall we continue this discussion in the Alternative Economic theory thread? This thread is just an illustration of that phenomenum.
What's fucking funny about it ?
What's not correct ?
That science was commercialized ?
That funding for fundamental research was all but stopped ?
That Wall Street attracted the laid off people ?
That these people made probabilistic theories ?
That the decisions makers, the bankers, abused those ?
WHAT is funny ? Make an argument instead of insulting me again.

[Michael]

What's fucking funny about it ?
What's not correct ?
That science was commercialized ?
That funding for fundamental research was all but stopped ?
That Wall Street attracted the laid off people ?
That these people made probabilistic theories ?
That the decisions makers, the bankers, abused those ?
WHAT is funny ? Make an argument instead of insulting me again.
There is nothing to discuss or argue about in this thread. I posted it only for the irony regarding the discussion in the other thread.

I apologize if you don't see the humor here.

The humor is that this is a perfect example of economics doing EXACTLY what you are criticizing them for NOT doing in the other thread (not using rigorous mathematical-based theories).

And it illustrates EXACTLY the point that I've been making all along that funky math and economics is a fool's game and doomed to fail.

The irony of that amuses me.