Re: st: about residuals and coefficients

[David Hoaglin <dchoaglin@gmail.com>]

Yuval,

I am grateful for Greene's textbook.  Unfortunately, however, it is
not difficult to find textbooks (in various fields) that interpret an
estimated coefficient in a multiple regression as the change in Y when
that predictor variable is increased by one unit and the other
predictors are held constant.  The authors of those books apparently
do not realize that, as the quotation from Greene shows, such an
interpretation does not reflect the way multiple regression works.

David Hoaglin

On Thu, Sep 5, 2013 at 2:55 AM, Yuval Arbel <yuval.arbel@gmail.com> wrote:
> I also checked in Greene's textbook - which is considered to be (at
> least by economists) the most extensive econometric textbook - which
> covers almost everything. On page 73 he specifically mentions the
> Frisch-Waugh (1933)-Lovell (1963) Theorem as Theorem 3.2.
>
> "In the linear least squares regression of vector y on two sets of
> variables, X1 and X2, the subvector b2 is the set of coefficients
> obtained when the residuals from a regression of y on X1 alone are
> regressed on the set of residuals obtained when each column of X2 is
> regressed on X1"
>
> P.S. It is possible that someone does not get the credit he or she
> deserves, but this is a matter for history researchers. A very famous
> example is the L'Hopital Rule, named after the wrong person. The story
> says that L'Hopital bought the right to be named as the original
> discoverer of the rule
>
> Reference: William H. Greene: Econometric Analysis, Seventh Edition
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/