Re: Experienced Statistician to help decide whether a regression is legitimate


David A. Heiser wrote:
"Anahita" <sputzele@xxxxxxx> wrote in message

To reply to some other poster:
-----------------------------
I cannot walk away from this mess (alas and even if I'd very much like
to).
More I will have to propose something potentially improving (...sic).


I'll settle to "less (pseudo) sophisticated treatment" but more
sensible...for instance a few well constructed descriptive statistics...
at least it will oblige them to sit down and think and maybe then they
will come up with something better..
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

This is a difficult question to answer, since the whole problem is
"proprietary" in nature, and you can only talk in very general verbal (and
confusing) constructs. As Bob says, hire somebody like OMU to help you out.

Hey, what about ME? :-) Actually big corporations had hired me
as consultant when several hundred millions USD were at stake, or
even when much less stakes were involved. :-) I didn't do much
consulting (even at high stakes) because I considered my profession
as statistical EDUCATOR was far more important. The fact that I am
giving FREE consulting (and getting annoyed and abused by the
malpractice posters <G> ) doesn't mean that I am not a highly
qualified consultant on many (certainly not all) APPLIED statistical
problems.

If "mathematical statistics" consulting jobs were offered to me, I
would probably pass them to Herman Rubin, or Jack Tomsky, or
some of the mathematicians or math statisticians who seldom
venture into the trenches of sci.stat.math/ed/consult. :-)

Obviously the companies have great "gobs" of cash available for this. Its
just a matter of spending dollars where it is important, and most older
companies are really, really bad at where they put there money. There is too
much management ego and management salaries that totally blocks effective
progress. They are like Ford and General Motors.

David Heiser

As a matter of fact, when I was talking about Harry Roberts and the
fact
that some BIG corporations REQUIRE their top management personnel
to have at least several courses in advanced statistics, I was thinking
that both Ford and General Motors were among those big corporations!

-- Reef Fish Bob.

.

Relevant Pages

  • Re: Sobering and annoying all at the same time....
    ... David Napierkowski ... The majority of those vocal act as though they can prove these links, and the honest truth is the science either is just not there and often cannot prove the link. ... Couple that with the extreme rise in lung-cancer rates (during the same period of a serious decline in smoking) in the US, and we're starting to see something important here. ... and the statistics behind that are much stronger than the statistics behind smoking. ...
    (alt.smokers.pipes)
  • The David Zatz FAQ, Part 1 of 1`
    ... David Zatz, allpar webmaster ... Background - Dr. David Zatz ... David Zatz currently works with Toolpack Consulting in Teaneck, ...
    (rec.autos.makers.chrysler)
  • The David Zatz FAQ, Part 1 of 1
    ... David Zatz, allpar webmaster ... Background - Dr. David Zatz ... David Zatz currently works with Toolpack Consulting in Teaneck, ...
    (rec.autos.makers.chrysler)
  • The David Zatz FAQ, Part 1 (was: rec.autos.makers.chrysler FAQ)
    ... David Zatz, allpar webmaster ... David Zatz currently works with Toolpack Consulting in Teaneck, ...
    (rec.autos.makers.chrysler)
  • Re: Birkhoff Causation is Consistent With Rare Events But Not With Others
    ... statistics is moving in a lot of directions. ... it's going in many directions. ... Of course Heiser has the germ of an idea in that there is enormous ... the differences between/among Rare Events, Fairly Frequent Events, ...
    (sci.stat.math)