Re: poly: Modeling Economic Singularities

From: Peter C. McCluskey <>
Date: Sun Apr 26 1998 - 10:50:28 PDT ("Perry E. Metzger") writes:
>Robin Hanson writes:
>> Perry responds to me:
>> >Hey, you made an extraordinary claim: that a short set of very well
>> >behaved (largely linear, as I recall) equations could tell us
>> >considerable useful information about the state of the economy into
>> >the far future. Extraordinary claims require extraordinary evidence.
>> All standard results in all fields consistitute extraordinary claims
>> to those who don't know those fields.
>But I *do* know the field. Not as well as you, admittedly, but I know
>it well enough to know 1) that similar models have never been shown to
>have significant predictive power

 I find this claim of yours a good deal more extraordinary than most
of Robin's claims.
 I think you are claiming one of the following things:
 a) Robin's conclusions are sensitive to quantitative inputs to his models
  whose values we have no theoretical reason to expect will remain within
  the assumed ranges (my understanding of why econometric models are
 b) because Robin uses math-style language to communicate his ideas, his
  models share the problems of econometric models
 c) you know of some other weakness of econometric models that I haven't
 If you are claiming (a), I think you could easily identify some reasonable
alternative quantitative assumptions which produce conclusions different
from the ones Robin draws. If you are claiming (c), I'd like to hear what
that weakness is.

>You seem pretty intransigently uninterested in exploring the
>assumptions of your own field.

 I have always found Robin to be willing to explore any of his assumptions
if I argue for an alternative that is as clearly thought out as his.

> Hell, you wouldn't go so far as to make
>the concession (requested by someone else other than me!) of
>expressing the terms in your equations as english words rather than as
>single letters.

 It sounds like you aren't taking the trouble to distinguish between
communication style and the substance of his communication.

Peter McCluskey          | Critmail ( | Accept nothing less to archive your mailing list
Received on Sun Apr 26 17:54:20 1998

This archive was generated by hypermail 2.1.8 : Tue Mar 07 2006 - 14:45:30 PST