From: John Conover <john@email.johncon.com>
Subject: Re: Proposal for new "engineer.sci.economics"
Date: 30 Aug 1999 20:10:44 -0000
Dan Parker writes:
> conover@rahul.net wrote:
>
> > Dan Parker writes:
> > > conover@rahul.net wrote:
> > >
> > > > hbrashears@earthlink.net writes:
> > > > > On 28 Aug 1999 21:24:14 GMT, gchand4059@aol.com (George) wrote:
> > > > >
> > > > > >I get the distinct impression that formally trained economists do not like
> > > > > >engineers coming into their field.
> > > > >
> > > > > I wonder what an engineer would think of an economist telling them how
> > > > > to build a bridge?
> > >
> > > > I wonder what an economist would think about Markov/Wiener telling
> > > > them about high entropy processes.
> > >
> > > There was a discussion on this list not so long ago about entropy,
> > > and the research I did seems to suggest scientific entropy (as opposed
> > > to metaphorical entropy), is not a significant issue in human production
> > > activities. The sun is held to create so much entropy, that our efforts
> > > do not measurably impact the total amount. Greenhouse gases were
> > > shown to be not entropy, but pollution that blocked entropy from
> > > escaping into space.
> > >
> >
> > Oh, Dan, I was referring to information entropy, a la Claude Shannon,
> > from the information-theoretic disciplines.
> >
> > The entropy of the time series of a stock's value, or a nation's GDP,
> > has very high entropy, or uncertainty.
> >
> > So, it is difficult to reconcile the concepts of macroeconomics with
> > high entropy systems-how could economists be certain?
>
> Sounds like the same trouble Prigogine is have with his
> entropy predictive work, in that the butterfly effect makes
> the thing useless. I think much of the uncertainty of current
> economics is a case of something moving at another level
> than entropy, but nevermind. The following is a snip on some
> economics done on the information aspect of entropy in any case.
>
Hi Dan. Yea, but it does raise an epistemological issue in the
profession of economics: is prediction, (in high entropy systems, like
the GDP, for example,) even possible at all? If not, are we fooling
ourselves with the grand concepts and equations of macroeconomics
where we attempt to do so?
John
BTW, note that the butterfly effect implies only that long term,
(whatever that term means,) prediction is impossible-there is a short
term horizon of predictability, but beyond that, things degenerate
into a 50/50 crap shoot. Kind of like the weather. It is not clear
whether a national GDP is such a system-it might be just stochastic.
But then, if it is stochastic, one could make the statement that the
function of macroeconomics is to ascertain-through inductive
rationality-the Nash equilibrium of the system, and base state
equations on that.
--
John Conover, john@email.johncon.com, http://www.johncon.com/