Econometrics

[Paul Davidson <pdavidson@xxxxxxx>]

Thanks Cliff -- some great examples-- but you also
forgot  the "Time on The Cross"  econometric study
that empirically demonstratedd that the blacks under slavery were better
off than if they were free.!

And how about the Club of Rome's demonstration that within twenty-years
of their 1973 study, the world would exhaust its petroleum
resource?

  I call to your attention that I did an empirical study for the
Brookings Institution, at the same time (1973) -- published in BROOKINGS
PAPERS ON ECONOMIC ACTIVITY -- that explained why the conventional theory
of exhaustible resources required unrealistic assumptions and therefore
the Hotelling formula for oprimally allocating exhaustible resources was
faulty --  and then went on to used past data- -as descriptive of
the past-- and to cautiously suggest that within given constraints of
technology etc, what the future price of oil would be by the turn of the
century.  I do not hesitate to point out that my "price
suggestion" of approximately 8-10 dollars in 1973 dollars turned out
to be way to high!  So much for using the past as a guide to the
future even for something that requires an important knowledge of
geological processes and engineering techniques..

 
At 02:38 PM 10/10/02 -0400, you wrote:
> I think I would be inclined to agree with Paul
> about the utility of
> econometrics as a means of empirical observation save in very limited
> ways.
> That is not to argue against quantification per se. On the whole, I
> would
> argue the very best analysis of the Great Depression is Kindleberger's
> "The
> World in Depression". It is far superior to either Temin's or
> Friedman's
> works on the Great dDepression. Kindleberger did not use one
> econometric
> equation, while Friedman and Temin devoted great efforts to
> econometrics.
>
> Schumpeter and Rostow contributed a great deal to our knowleddge of
> leading
> sectors and industrialization. Econometricians came up with the
> brilliant
> observation that if we hadn't built railroads, we would have to
> build
> canals. If we hadn't developed east to west, we could have developed
> north
> to south (which actually ignores the obstacles to development in
> the
> southern states). Based on this incredible insight, they concluded we
> could
> have gotten along just fine without railroads...
>
> Post-Keynesians would do better to re-emphasize and rethink the role
> of
> historical narrative or is the proposition that history matters to
> be
> reduced to propositions about idempotent matrices?
>
> -----Original Message-----
> From: Sven R Larson
> [mailto:slarson@xxxxxxxxxxxx]
> Sent: Thursday, October 10, 2002 2:06 PM
> To: pkt@xxxxxxxxxxxxxxxx
> Subject: Fwd: Re: [Fwd: nobel]
>
>
>
> paul davidson wrote:
> > Don't confuse econometric research with mere observations!
> Most
> econometric
> > research MISUSES the statistical tools -- since the conditions
> necessary
> > for tests of significance, etc are not available for most
> economic
> > data.  Poor statistical analysis is bad statistics -- or as the
> book I
> read
> > when I was a biochemist years ago stated  in its title
> "How to Lie With
> > Statistics".
>
> Honestly, Paul, anyone with 300-level economics, and particular
> with
> graduate training, knows this. You're breaking through open doors.
> Even my son, 8 years old, knows that it makes no sense to say that
> if
> your right hand is in boiling water and your left hand is on ice
> you
> are on average doing pretty OK. So let's skip that and talk about
> how
> we can actually approach the world beyond the armchair. I have a
> review of Truman Bewley's book "Why Wages Don't Fall During a
> Recession" in the last ROPE. Read it, then read the book. Bewley
> shows
> that using the right kind of empirical tools you CAN grasp the
> world
> in a way that suits a Post Keynesian perspective.
>
> > Crude statistical analysis is bad statistics and deserves to be
> recognized
> > as such.  Observatiuon and statistical analysis is not the
> same.  As a
> > biochemist I was always warned that a column of data that others
> collected
> > was merely a column of numbers-- and not information about the real
> world
> [
> > [as those who have been following the recent several decade
> long
> > disclosures of fabrication of  lab book observations by
> eminent  and up
> > coming scientists has revealed.  and like Enron that is just
> the tip of
> the
> > iceberg in the natural science-- and econometrics starts even off
> worse,
> > since the lab book observations are fabricated by someone else who
> the
> > analyst typically does not know -- and does not communicate
> with.
>
> 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.
>
> /srl
>
> -- 
> Dr. Sven R Larson
> Department of Economics
> Skidmore College
> 815, North Broadway
> Saratoga Springs, NY 12866
> (518) 580-5278

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