Czwang writes:

> Can any one here suggest a good set of self-study statistics
> books for a computing scientist with only elementary
> probability and statistics knowledge (one undergraduate course 12
> years ago) so that he could master the book "The Elements of
> Statistical Learning"? or better, he could become a data miner using
> statistics learning methods. Thanks in advance. --czwang

You would need a lot of work to become good at data mining. I assume that
getting a masters degree in Statistics is out of the question. Too bad.
That's the sort of background you would need in order to understand the
basic principles of data mining.

If you had to start somewhere, you should look for good books on linear
regression analysis, logistic regression analysis, and multivariate
analysis. These topic areas are so broad that it is difficult to specify
what the best self-learning books would be. Also, it's hard to know what
book to recommend without knowing your tolerance for mathematical details.
The sorts of books that work well for someone comfortable with multivariable
calculus are quite different from the books that work well for someone who
is frightened by high school algebra.

You probably also need to learn a good software package. R is an excellent
open source package, and its object-oriented approach will be a natural for
someone with your background.

If you plan to program these data mining algorithms yourself, then invest in
a good book on statistical computing.

You might try searching the archives of sci.stat.edu and other newsgroups
for textbook recommendations in some or all of the above areas.

It won't be easy. This is not something you will be qualified for after
reading one or two books. But if you do decide to follow this career path,
it should be quite interesting and exciting. There's a lot of demand for
data analysis skills and the work is far more engaging than most programming
jobs, in my humble opinion.

Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
The STATS web page has moved to
http://www.childrens-mercy.org/stats.

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to