[OPE-L:6192] Re: econometrics

From: Gerald_A_Levy (Gerald_A_Levy@email.msn.com)
Date: Sun Nov 18 2001 - 07:19:20 EST

Re [6191]:

Hi Jurriaan. It's good to hear from you.

> Insofar as econometrics is concerned with the measurement of economic
> trends in order to build models and extrapolate future trends (make
> predictions), this is surely a worthwhile pursuit for the followers of
> Marx. <snip, JL>
> The point of economic
> research is rather to bring theory and the data closer together, so that
> theory is disciplined by the data of experience, and the study of the data
> is informed and guided by theory. Econometric techniques are certainly
> useful in this regard ! <snip, JL>

1) I reject the overly broad description that you give above [in the first
sentence] for econometrics.

2) You have risen to the defense of what I, and some others, might call
"Quantitative Marxism"  Yet, I have not questioned the need for quantitative
empirical analysis by Marxists -- I have questioned the use of econometrics
by Marxists. Thus your reply does not seem to me to be very responsive
to the issues I raised.

3) Some non-econometric quantitative methods include input-output
analysis, game theory, and chaos theory. This, of course, does not suggest
that there are not problems with those methods -- but the point that I
want to make here is that they represent *alternative* methods for
conducting empirical analysis. Indeed one might argue that they can
represent *preferable* methods (e.g. in the article by J. Coakley
in *Quantitative Marxism* after offering a critique of econometric models
regarding stock market efficiency, the author concludes not by offering
an alternative econometric formalization but rather by claiming that
 "...only the chaos theory approach and, to a lesser extent, the speculative
bubbles model appear to capture these features within a quantitative
framework. Marxian economists should take up these issues and begin to
challenge the hegemony of orthodox theories of financial markets", p. 122).

3) Both micro and macro econometric studies have employed, amongst
other absurdities, factor (of production) analysis (and all that implies
regarding factor endowments and productivity) and production functions
(imagine the audacity of anyone employing an aggregate production
function after 1960!).  My question concerns whether econometrics,
which is after all a branch of neo-neoclassical analysis, is necessarily
wedded to or can be divorced from marginalism.

In solidarity, Jerry

Paul Dunne *Quantitative Marxism*,  London,  Polity Press, 1991
[ISBN 0 -7456 - 0647 - 4], Out-of-print

This archive was generated by hypermail 2b30 : Sun Dec 02 2001 - 00:00:06 EST