File spoon-archives/bhaskar.archive/bhaskar_1998/bhaskar.9809, message 29


From: "Tony Lawson" <TL27-AT-econ.cam.ac.uk>
Date:          Tue, 22 Sep 1998 13:07:16 BST
Subject:       Re: BHA: Regression paper 


Hi Doug

I worry about venturing out of my position of permanent lurker to this 
list, but I wonder if you could clarify for me which of the 
following positions, if either, you are defending:

1) regression analysis does not require a closure for the "parameter 
estimates", etc., to be meaningful/interpretable. 

2) regression analysis only requires a closure within the sample 
period or domain.

If 1) I would be interested in how this conclusion is reached; if 
2) I would be interested in wether you think this will get us far in 
social world. 

I too enjoyed your talk very much, and would like to see a copy of 
the paper.

Very best

Tony 






> Date sent:      Mon, 21 Sep 1998 12:14:02 -0400
> To:             bhaskar-AT-lists.village.virginia.edu
> From:           Doug Porpora <porporad-AT-duvm.ocs.drexel.edu>
> Subject:        BHA: Regression paper 
> Send reply to:  bhaskar-AT-lists.village.virginia.edu

> Hi Everyone,
> 
> Nick, thanks for your interest in my paper.  Andy, thanks for that synopsis
> of my paper and the ensuing discussion.  I just need to type in some of the
> changes I made while on the road and then will send you both a copy in Rich
> Text Format. Marshall, I seem to recall your wanting a copy as well and
> haven't forgotten.  I also intend to send a copy to CCR in time for posting
> all the papers on the Web site.
> 
> Andy, you're right, I'm in sociology, which, it surprises me to say, does
> not seem quite as bad as econometrics.  I would make only one qualification
> in your description of my position.  I think the crux of the matter between
> me and the CR economists is that I do not think that regression is an
> explanatory tool at all but only an evidentiary too.  Thus, although I
> quite agree with the CR economists that we are unlikely to find any
> time-universal conditions of closure necessary for regression to be a
> nomothetic explanation, such time-universal conditions of closure are
> unnecessary for regression _often_ to be one useful piece of _evidence_ for
> the contingent operation of a mechanism at one sociohistorical point and
> time. Stable conditions are unnecessary if you are not positing a
> time-invariant explanation.
> 
> Andy, I really enjoyed meeting you, too, and hope we can continue to be in
> touch on these issues.  More generally, I very much enjoyed seeing the
> people I already knew, like Colin and Caroline, the people I knew only
> virtually, like Gary, and the many new people I met.
> 
> For those of you who so far have not been able to attend the CR conference,
> I can only say it was great -- and amusing as well -- for me to be at a
> conference where so many people shared so many of my basic sensibilities
> that I could even appear to be the positivist in the bunch.  More
> seriously, at a conference with those with whom you share a framework, you
> are able to explore higher level issues instead of fighting and refighting
> the basics.  For me that was novel and refreshing.
> 
> doug
> 
> doug porpora
> dept of psych and sociology
> drexel university
> phila pa 19104
> USA
> 
> porporad-AT-duvm.ocs.drexel.edu
> 
> 
> 
> 
>      --- from list bhaskar-AT-lists.village.virginia.edu ---
> 


     --- from list bhaskar-AT-lists.village.virginia.edu ---

   

Driftline Main Page

 

Display software: ArchTracker © Malgosia Askanas, 2000-2005