CAMREC: Re: Turing tests?

From: the Campaign for Real Economics (camrec@mmu.ac.uk)
Date: Fri Jul 09 1999 - 16:21:51 BST


Date: Fri, 09 Jul 1999 16:21:51 +0100
Message-Id: <199907091521.QAA26978@alpheratz.cpm.aca.mmu.ac.uk>
To: bogus
From: camrec@mmu.ac.uk (the Campaign for Real Economics)
Subject: CAMREC: Re: Turing tests?
In-Reply-To: <Re: Turing tests?>


Message-Id: <199907091521.QAA26978@alpheratz.cpm.aca.mmu.ac.uk>
To: CAMREC list members <camrec@mmu.ac.uk>
From: the Campaign for Real Economics <camrec@mmu.ac.uk>
Date: Fri, 09 Jul 1999 16:21:51 +0100
Subject: CAMREC: Re: Turing tests?

Scott wrote:
> The proposal for an economic Turing test (ETT) is an an interesting
> approach to undermining conventional economics, but I wonder if it would
> be a diversion from the more important objective of devising approaches
> to economic analysis that would be more relevant to policy analysis.
>
> An alternative (though it could happen in parallel with the ETT) would
> be to identify a small number of economic problems with important policy
> implications that have either been ignored by the economics profession
> or have been badly handled. Perhaps a problem arising in the emerging
> market economies or a comparison of different regimes with apparently
> different experiences of economic success -- say Hungary or Poland and
> the Russian Federation. Another problem might relate to issues
> important in (say) Sub-Saharan Africa. Another might relate to the
> functioning of real markets since economists -- bizarrely -- reject
> institutionally realistic representations of markets as "anecdotal".
> Another old-fashioned question that has been assumed away by economists
> is the limits on firm size or growth rates. Transactions economists
> fudge the issue by appealling to some Heaven-sent menu of organizational
> arrangements for the firm.
>
> My preference would be for the development of analyses that start with
> the problem. Agent representations should be appropriate to the problem
> rather than specifying problems so that they can be addressed with
> preconceived agent representations (such as "rational" optimizing
> agents).

I agree that ultimately unreal economics will fail due to the paucity of
its models, and its ability to substantially ignore the real world. I
also agree that the way forward is models that take as much from real
problems/systems and use as few a priori assumptions as possible.

However experience suggests that economists will always quibble as to
how successful they are at addressing real problems, which would lead
one back to the old unproductive aguments about why what they do is bad
science. External observers will be confused by the competing claims
about the success of these traditional models - the failure of economics
will not be made apparent.

The ETT forces evaluation only by its success at replicating the
behaviour of a real economic entity. We can choose a realistic problem
domain for the interaction and thus force entrants to deal with it (and
not an abstraction of it).

Regards.

--------------------------------------------------
Bruce Edmonds,
Centre for Policy Modelling,
Manchester Metropolitan University, Aytoun Bldg.,
Aytoun St., Manchester, M1 3GH. UK.
Tel: +44 161 247 6479 Fax: +44 161 247 6802
http://www.cpm.mmu.ac.uk/~bruce



This archive was generated by hypermail 2b29 : Thu Jan 06 2000 - 17:00:31 GMT