econometrics is not really science as much as it's the tool used by
economists to help predict future behaviour and future value... Schumpeter
said: "It is unreasonable to expect the economist to forecast correctly what
will actually happen as it would be to expect a doctor to prognosticate when
his patient will be the victim of a railroad accident and how this will
affect his state of health". perhaps less cynical an answer: economists
(with or without econometrics) can't exactly model real world behaviour
through lab experiments. if they were able to, i'd be out of a job and
there would be no need for markets because markets would be predictable.
phillip
btw, yes, brokerage firms did try to employ mathematicians in the late '80s
and early '90s to bring some understanding and hopefully create predictors
of momentum, value, etc. the experiment didn't work and most of the
mathematicians were eventually sent back to university -- so much for
applying scientific analysis/discipline to the economics of markets.
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]]On Behalf Of Jim Choate
Sent: Monday, April 23, 2001 6:38 PM
To: [EMAIL PROTECTED]
Subject: Re: The Well-Read Cypherpunk
On Tue, 24 Apr 2001, Faustine wrote:
> What about econometrics? It seems to belong in the same conceptual
category
> as mathematics, statistics, operations research, etc. I dont think
> econometricians would generally appreciate being called softies...
If by 'econometrics' you mean 'take real world data and try to fit models
to it' then I'd say it clearly is a science. A science is the process, not
the thing it's applied to. It's a way of viewing the cosmos. It has all the
requirements including repeatability and predictability, neither of which
'economics' has. I like derivative junkies. If on the other hand you mean
'take your pet theory and munge data until it fits', no thanks. THAT's the
vast majority of economics and politics. Because of the very nature of
politics there is no way economics will ever mature until it breaks those
ties (one aspect of the Anarcho-* paradigm Democratists share, at some
point the economics of a society must seperate from the ethics and
practice of 'government' - though I strenously disagree on their
approach).
An example would be the application of diffusion equations to money
distribution in various sorts of markets. I've never seen a 'economics'
text derive that relation or predict it, the reaction of every economist
and the literature when it became clear the application worked was
'imagine that'.
I'd posit that through 'applied science' to economics over the last 20-30
years we've made more progress in understanding the concept of market than
we did in the previous 5,000.
Take a look at Mandelbrot's recent series of books on economics/econometrics
and chaos theory. The only one I've even looked at is the 1/f noise one
however.
> I think any economic theory worth its salt requires good data obtained by
> some kind of scientific methodology. Some economic theories are more
> scientific than others, it all depends on the approach.
They not only have to get 'hard' data, but 'soft' data in regards the
motives of the actors in the markets. One of my major objections to
'economics' is that it completely ignores this aspect (ie rational). Any
economic theory that does not axiomatically include human psychology is
incomplete. The result of this is the elimination of 'rational' in
economic theory and instead we inject some statistical term to replace it.
Really my objection to 'economics' isn't so much with it per se, but
rather with the variety of 'card sharps' who try to use it to take
advantage of people by promising more than they can deliver (ie politics
& profit).
> Kind of reminds me of this nuclear physicist I knew who heaped scorn on
> economics as 'soft' every chance he got. But then, he also had
reservations
> about including the life sciences (like biology) as 'hard science' i.e,
> 'real' science. And heaven help the person who got him started on
psychology
> and sociology. I guess some people are just more hard core than others!
Boy, there are several issues that spring out in those few sentences...
By 'soft' I'd say a POTENTIAL science where even the fundamental axioms
are not well known. If one believes in the 'ghost in the machine' for
example then your physics friend is right. If the physics of our brains
are not sufficient for us to understand the cosmos then he's again right.
One thing is for sure, we'll never know unless we go there in person. I'm
a Pantheist, it's all one big cosmos. The question is can this little
corner of it understand itself? That's an open question. Psychology has
the POTENTIAL to become a science but only when we can build and design
brains, of whatever type, ourselves. I've some questions along that line
at the bottom of the URL,
http://einstein.ssz.com/ravage/pantheism.html
Life sciences is becoming a hard science right before our eyes,
DNA Computing: New computing paradigms
G. Paun, G. Rozenberg, A. Salomaa
ISBN 3-540-64196-3
Algorithms On Strings, Trees, and Sequences: Computer Science and
Computational Biology
D. Gusfield
ISBN 0-521-58519-8
It's one of the reasons I'm so intrigued with it. It's like being alive in
Giordano Bruno's day. (Though I actually suspect it's always like this
now/then basically, just different toys to play with)
The only way that sociology will EVER become a science will be when we
meet another alien species AND we can build brains of whatever design we
desire (and possibly Ecology with respect to xeno-forming, w/ Biology a
science humankind is no longer constrained by our current biology - See
Dougal Dixon's "Man after Man"). Then perhaps...(hint: Cliology)
____________________________________________________________________
The solution lies in the heart of humankind.
Chris Lawson
The Armadillo Group ,::////;::-. James Choate
Austin, Tx /:'///// ``::>/|/ [EMAIL PROTECTED]
www.ssz.com .', |||| `/( e\ 512-451-7087
-====~~mm-'`-```-mm --'-
--------------------------------------------------------------------
