[OPE] Gillian Tett on the magic of models

From: Jurriaan Bendien <jurriaanbendien@online.nl>
Date: Sun Dec 26 2010 - 11:05:58 EST

How not to fall foul of the model makers
By Gillian Tett

Published: FT December 23 2010

(...) When the financial crisis hit, many observers blamed the disaster on
the misuse of financial models. Not only had these flashy computer systems
failed to forecast behaviour in the sub-prime mortgage world, but they had
also seduced bankers and investors to take foolhardy risks or been used to
justify some crazy behaviour. But these days, in spite of all those
mis-steps, there is little sign that financiers are falling out of love with
those models; on the contrary, if you flick through the recent plethora of
reports from the Basel Committees - or look at the 2011 forecasts emanating
from investment banks - these remain heavily reliant on ever-more complex
forms of modelling.
So what are investors to make of this? One particularly thought-provoking
set of ideas can be seen in the current work of Emanuel Derman (...) As he
explains in a recent paper and forthcoming book, in the world of physics,
scientists can deduce theories, laws and equations which tell you how the
world works in its own, absolute terms. But "there are no genuine theories
in finance because finance is concerned with value, an even more subjective
concept than heat or pressure," he observes. And unlike physics, financial
experiments are not repeatable - history changes how markets work.

Thus while investors might like to see models as akin to physics theories,
Derman says this is wrong; models are more like expansive metaphors, of the
sort found in literature or philosophy. In essence, they are a tool to help
us think, and order our world view, and explain something that is hard to
grasp. "Models are reductions in dimensionality that always simplify and
sweep dirt under the carpet. Theories tell you what something is. Models
tell you merely what something is partially like."

But that does not, Derman adds, mean that models should be junked. They can
be valuable if employed with five, key principles: firstly, financial models
should avoid too much axiomatisation; secondly, they should be as "vulgar"
as possible (meaning there should be "as direct a path as possible between
observation of similarities and the consequences"). Thirdly - and
crucially - they should "sweep dirt under the carpet but let users know
about it" (ie, tell everyone up front what has been ignored in the model
assumptions.) Fourthly, model users should be trained to think of models as
"Gedanken experiments", used to illustrate a theory but never meant to be
physically carried out - like Schrödinger's cat. Last but not least,
financial players must stop their "idolatry" of models; and, presumably,
their worship of quants. (...) But, sadly, there is still plenty of
"idolatry" about...

Full article

ope mailing list
Received on Sun Dec 26 11:07:33 2010

This archive was generated by hypermail 2.1.8 : Fri Dec 31 2010 - 00:00:02 EST