In article <[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
>Dino, I'm not an expert, I come here for advise, but certainly I would
>think any statistician would need to understand probability first,
You would be surprised at how many statisticians have little
understanding of probability, and how many of those who have
more consider it as formal, and only look at calculating
probabilities. The important part is understanding probabilities,
which seems almost independent of calculating.
>then univariate and multivariate statistics, as a foundation.
It is really necessary to start much farther back, with the
idea of a decision problem. The fundamental problem can be
stated in one sentence:
It is necessary to simultaneously consider all
consequences of the proposed action in all
states of nature.
Much of current statistical practice would be excluded
by application of this simple principle. The principle
does not, by itself, tell what to do, but it states that
the question needs to be asked.
If you
>will be able to maintain access to it, SPSS or SAS are among the more
>powerful tools to know, and learning them will force you to understand
>the stats.
This is false. These are heavily used by disciples of
various cults of the statistical religion, who use the
methods as mandated divinely, but without knowing what
they are doing, or why.
Learn concepts and theory; correct application is then
easy. Learn formulas and methods, and it becomes
harder to achieve the understanding.
Seems like around this campus, SAS is what the hardcore
>stat folks use, but you have to know what you are doing. HTH
>Jim
>On Sun, 06 Jan 2002 18:21:13 +0800, Dino Hsu
><[EMAIL PROTECTED]> wrote:
>>Dear all,
>>I am a programmer and database administrator, I have been equipped
>>with IT basics. After I do OLAP (data mart) and data mining, I am
>>turning into a statistics + IT cross-disciplinary area for master and
>>doctor degrees. Can anyone tell me what the basics of statistics are
>>required? 1. Introduction to Statistics 2. Multi-variable analysis? 3.
>>statistics tools (statistica?) Anything else?
>>Thanks,
>>Dino
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
[EMAIL PROTECTED] Phone: (765)494-6054 FAX: (765)494-0558
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
