Re: [Fwd: nobel]

[Paul Davidson <pdavidson@xxxxxxx>]

At 02:05 PM 10/10/02 -0400, you wrote:

> Are you saying qualitative evidence is
> superior to quantitative? I
> don't think so, but you don't seem to have anything good to say
> about
> quantitative empirical methods whatsoever.


Maybe that is because I was trained as a biometrician and know that
before anyone attempts to disprove conjectures, they must have an
experimental design which assures that the universe is  under
statistical control.  At best, honest data collection and analysis
(i.e., that  pass some analysis of variance test) of economic data
(times series?) can be descriptive of the past HISTORY -- 

and historical evidence, as even Paul Samuelson has recognized (see his
Canadian Journal of Economics 1969 article), is not sufficient for
"scientific" analysis unless one presumes, what Samuelson
called "the ergodic hypothesis". 

Since I reject the ergodic hypothesis for all important time series 
variable analysis-- I can not say any good things about such an
analysis   But, without always realizing the implication of
what they are saying the findings of many mainstream economists , e.g.,
Christiano and Eichenbaum [ "Unit Roots in GNP? Do We Know and Do We
Care?" CARNEGIE-ROCHESTER CONFERENCE ON PUBLIC POLICY, VOL. 32,
edited by A. Meltzer], indicate that econometric analysis of past time
series data may--at best be descriptive of historical events-- but that
is all.

Sorry Sven --

Paul

Paul Davidson

Editor, JOURNAL OF POST KEYNESIAN ECONOMICS

Economics Department - University of Tennessee

503 SMC

Knoxville, Tennessee 37996-0550

work phone: (865) 974-4221

fax: (865) 974-4601/  (865) 974-1686

home phone and fax (865) 692-0802

http://econ.bus.utk.edu/davidsonextra/Davidson.html