A curious thing, or so it seemed at the time, was a statement Joe Hodges made to a nonparametrics class, "The ideal mathematics lecture would not involve any writing on the blackboard." In an era when, as you note, one need simply pull down a menu and select t-test, it seems instructors of statistics ought to be getting closer and closer to Hodge's goal. Freedman, Pisani, Purves, and Adhikari has only two formula in the first four chapters and pretty much sticks to that formula density throughout.
Donald Burrill <[EMAIL PROTECTED]> wrote:
Donald Burrill <[EMAIL PROTECTED]> wrote:
Reply to OP and the list.
There used to be a useful book entitled "Computational Handbook of
Statistics", by Bruning and Kintz; don't recall the publisher.
Original edition was published about 1970 and I was aware of at least a
third edition some years later; don't know if it's still current, but
copies might be found in a university library. Probably viewed as not
useful these days, when if you want a t-test you pull down the "stats"
menu in a software program, select "t-test", and follow the dialogue
boxes, and don't have to do the arithmetic "by hand".
It was devoted to describing how to carry out standard statistical
techniques; it sort of treated the reader like a computer terminal, and
programmed the reader through each procedure. ("Step 1. List the data
[showing a useful format for so doing]. Step 2. Add the data in each
c! olumn." And so on.) For each technique that it described it also
provided a short list of references, which the reader could consult if
he wanted to know anything about the statistical theory underlying the
technique.
I don't know the Bronstein book, so don't know whether Bruning & Kintz
might be useful to you or not.
On Sat, 31 Jan 2004, enclume42 wrote:
> I am looking for a handbook of statistics a bit similar in style to
> that the Bronstein has for general mathematics.
>
> I have a physics background, thus I am familiar with math, but I am
> missing some statistics terminology and I miss some basic statistical
> methods. Ideally the book should thus cover all what a good graduate
> statistician should be aware of.
>
> For example, I would like to find each classical variant of a t-test
> described in ~2 pages, with a short introductory text describing the
> ideas behind the metho! d, followed by a list of the assumptions of the
> method (e.g. normality of data distribution) and then the procedure to
> actually conduct the calculations (in mathematical terms). Examples
> and references to specific statistical packages should be kept to a
> minimum.
>
> I have been looking for such a book for months, but I always end up
> finding textbooks in which the essence of the methods is diluted into
> pages of general -even trivial- considerations (which makes me fall
> asleep), or in specialized books in which the basics are missing (the
> book should start with the definition of the mean).
>
> Any suggestion ? :-)
Bonne chance! -- DFB.
------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
56 Sebbins Pond Drive, Bedford, NH 03110 (603) 626-0816
.
.
=================================================================
Instructions ! for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
Phillip Good
http.ms//www.statistician.usa
"Never trust anything that can think for itself if you can't see where it keeps its brain." JKR
http.ms//www.statistician.usa
"Never trust anything that can think for itself if you can't see where it keeps its brain." JKR
Do you Yahoo!?
Yahoo! Mail SpamGuard - Read only the mail you want.
