On Sat, 22 Jul 2000, Christopher Tong wrote:
> Does anyone have recommendations for introductory
> books on regression analysis? I posted this question on
> sci.stat.math and got only one reply so far.
Depends on where you're coming from and where you want to go, inter alia.
Your e-mail address suggests your background may be in physical sciences,
and a text written by, or for, physical scientists might be most useful.
That puts me in mind of Draper and Smith; the book was originally
written at the behest of the Chemical Division of the ASQC, and its
numerical examples are from industrial chemistry.
However, there is some evidence that in statistics (perhaps more
than in most disciplines) there is a strong interaction between writing
style and reading style, especially at introductory levels; and perhaps
your best strategy would be to immerse yourself for a time in your
university library, reviewing a fair spectrum of books that deal with
multiple regression, and taking home the ones you find most eminently
readable.
I am not personally familiar with Neter et al., but have heard
good things about it from people whose opinion I value.
Books that I have found useful, and consider to be very good for
their own purposes and from their own viewpoints, include (in no special
order) Bottenberg & Ward, Darlington, Judd & McClelland, Pedhazur,
Edwards, and Cohen & Cohen; the latter five have fairly heavy emphases
on psychological and/or educational substantive areas, from which they
draw most of their numerical examples. (I was teaching at OISE, a
graduate school of education, for which such examples were appropriate.)
> I am currently using Neter, Kutner, Nachtsheim, and
> Wasserman, which I find unwieldy and not very concise.
"Not concise" is not necessarily bad. When one understands the stuff,
concise is preferable; but when one doesn't, more verbose textual style
can be helpful. I suspect there's a distinction in there somewhere
between a style that's useful and helpful to learn from, and a style that
facilitates retrieval (as it were from a reference) after the material
has been well learned. I remember seeing some years ago a report of
research on statistics teaching (sorry, can't recall details to cite) in
which the textbooks preferred by the students were distinctly NOT the
textbooks preferred by the instructors, and the students' preferences
were rather strongly inclined toward the more verbose textbooks.
> I have my eye on Montgomery & Peck, but am wondering what anyone
> else would recommend. My one reply so far suggested Cohen & Cohen.
------------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED]
MSC #29, Plymouth, NH 03264 603-535-2597
184 Nashua Road, Bedford, NH 03110 603-471-7128
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
