Re: Duncans work on statistical mechanics and economics

From: clyder@GN.APC.ORG
Date: Fri Jun 25 2004 - 08:29:54 EDT


I am just rather uneasy about having each agent as
a thermodynmaic system since I have doubts about the
reality of the utility schedules on which it is based
and also to me entropy seems so tied up with integrals
over probabilities that I have difficulty equating
it with utility

Quoting Philip Dunn <pscumnud@DIRCON.CO.UK>:

> Quoting Paul C <clyder@GN.APC.ORG>:
>
> > Philip Dunn wrote:
> >
> > >Quoting clyder@GN.APC.ORG:
> > >
> > >
> > >
> > >>Has anyone else here read Duncan Foley's work on statistical
> > >>mechanics and neo-classical economics?
> > >>
> > >>----------------------------------------------------------------
> > >>This message was sent using IMP, the Internet Messaging Program.
> > >>
> > >>
> > >>
> > >
> > >Hi Paul
> > >
> > >Not the statistical mechanics, only:
> > >
> > >Classical thermodynamics and economic general equilibrium theory (with
> Eric
> > >Smith)
> > >
> > >
> > >Philip Dunn
> > >
> > >
> > >
> > >
> > I realise that I was wrong to say statistical mechanics. It is very
> > much classical thermodynamics whereas my familiarity is more
> > with the statistical mechanics or Shannon concepts of entropy.
> >
> > I thus find Duncans identification of entropy with utility of a single
> > agent very
> > strange since to me the idea of entropy has to be based on
> > summing over large ensembles.
> >
>
> My first thought would also have been 'agent=atom in a gas' but DF takes
> each
> agent as a thermodynamic system.  I don't think statistical mechanics is
> behind
> this. I was impressed by the way that the GE initial endowments and trading
> to
> equilibrium were dropped in favour of reversible near-equilibrium trading
> (keeping the refrigator stocked). Rather like Marxian reproduction, in fact.
>
>
> Philip Dunn
>
>




----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.


This archive was generated by hypermail 2.1.5 : Wed Jun 30 2004 - 00:00:02 EDT