Nassim Taleb Thinks Everyone Is Stupid

BNET TalkbackShare your ideas and expertise on this topic

Subscribe to this discussion via Email or RSS

•  
  1

  Joe.enlight

  05/15/08 | Report as spam

  Goldman's Alpha Fund

  Nassim is talking science - you're talking anecdotes. Every major bank and most hedge funds have algorithmic-trading funds (Goldman's is worth $6B) that trades solely electronically without human interaction. The overuse of models that most MBAs don't understand is a real issue. Goldman's Alpha fund dropped 20% in September 2007 alone - mostly because humans don't interact with it.

•  
  2

  Michael Fitzgerald

  05/16/08 | Report as spam

  RE: Nassim Taleb Thinks Everyone Is Stupid

  Nice line (science vs. anecdote). But really, we had a similar issue with program trading in the 80s. If we're still having the issue, you can't blame the computers. Or are you saying that the financial system is so dependent on its technology-based models that people non longer matter?
  
  Michael Fitzgerald

•  
  3

  Joe.enlight

  05/16/08 | Report as spam

  People vs. Computers

  That is precisely the issue - they didn't learn anything from the '80s. The only takeaway was that "we need better models." The rise of IT has assisted in this mindframe. Traders now buy building to put their servers closer to the electronic exchanges to improve execution time - we're talking speed of light here.
  
  Too many banks relied on quantitative risk models last summer. They had no clue what their real risks were, just their Daily VaR.
  
  My favorite line (not sure to whom to attribute it) - "Why do banks come up with so many new ways to lose money when the old ways work perfectly well."
  
  I'll agree that Taleb is a bit full of himself, but he communicates the central idea so much better than Mandelbrot, the real brains behind the operation.

•  
  4

  Michael Fitzgerald

  05/18/08 | Report as spam

  People and computers

  I thought that's what you were saying. So do you think that the balance can shift, so that the computers are tools that inform our judgment, rather than making (well, being programmed to make) the judgments for us?
  
  Or have we reached that state of being helpless without the technology that some philosopher of science (I forget which one) says goes with any technological development?
  
  btw, I don't personally think Taleb is wrong to say the [model] emperor has no clothes on. What surprises me is he doesn't seem to be getting much argument.
  
  Michael Fitzgerald

•  
  5

  Joe.enlight

  05/20/08 | Report as spam

  Taleb's Case

  What can his opponents say? They likely know the math is wrong. It is hard to argue otherwise. Most of the founders of this method have had their lunch handed to them - think LTCM.
  
  NNT had great timing in the release of his book. He uses credit-rating agencies as an example of things proponents bring up to "prove" that the methods work. One month later the agencies fall apart.
  
  I watched the credit crises evolve from the Treasury department of a super-regional bank. Everything we did was based on the normal curve. It is simply ingrained into everything we do. We made rate prdeictions (normal curve) to assess changes in the value of assets (more normal curve). If you told us last June that we wouldn't be able to securitize and sell for funding, you would have been laughed out of the room.
  
  The most successful banks and hedge funds have been those who thought about risk holisticly (Goldman did a great job, despite the Alpha Fund). Those that were over leveraged to exploit small modeled advantages (WM, NCC, BSC) were hugely profitable when it worked, but are destined for the dustbin.
  
  Thanks for the exchange. Very enjoyable.

•  
  6

  Michael Fitzgerald

  05/20/08 | Report as spam

  NTT

  The pleasure was mine. Thank you.
  
  Michael

•  
  7

  Caspary

  06/23/08 | Report as spam

  RE: Nassim Taleb Thinks Everyone Is Stupid

  test

•  
  8

  siva1982

  05/04/09 | Report as spam

  Siva

  As Micheal says, the problem is not with the financial models
  but with the users. Financial valuation was once considered
  an art. By using quantitative techniques we converted it
  meticulously into a science that could have universal
  application and consistent results. However every scientific
  theory has its limitations. Theory of Gravity always works on
  everyone. However it works differently on Mars than on
  Earth. Similarly, environment makes a lot of difference to a
  financial model. It would work if and only if the assumptions
  are realistic. A model will only serve the purpose for which it
  is built, it is not an omni-potent or some alien creature that
  can undergo mutations. It is necessary to understand the
  limitations to appreciate the usefulness. I found some really
  useful articles in http://www.financialmodel.net .

•  
  9

  Michael Fitzgerald

  05/04/09 | Report as spam

  RE: Nassim Taleb Thinks Everyone Is Stupid

  very nice line about how gravity works differently in different places.
  
  Thanks.
  
  Michael