3 upcoming short courses in statistics

Hello - The following courses in recent topics in statistics are
sponsored either by the Institute for Professional Education or by the
Gordian Institute. Details and signup information follow. Please
respond directly to the sponsor for more information.

(To be removed from this mailing list on statistical courses and
software, please reply to this message with "remove" in the

1. "Resampling, The New Statistics," Arlington, VA (1 subway stop
from Washington, DC), March 4-6, 1998

2. "Data Mining: Principles & Practice," January 28, 1998: Santa
Clara, CA. February 11, 1998: Satellite Beach, FL (Orlando Area)

3. "Advanced Techniques for the Analysis of Financial Markets,"
January 20, 1998, Herndon, VA (a suburb of Washington, DC)



1. "Resampling, The New Statistics," Arlington, VA (1 subway stop
from Washington, DC), March 4-6, 1998

Presenter: Peter C. Bruce, Vice President, Resampling Stats, Inc.

Resampling -- including the bootstrap, permutation, and other
non-parametric tests -- has taken statisticians by storm worldwide in
the past decade. Resampling is the method of choice for hypothesis
tests, confidence limits, and other applied everyday problems in
statistics and probability. It involves no formulas or tables, and
every step is perfectly understood. Resampling helps you avoid the
"Type 4" error of choosing the wrong procedure because you will
understand, step-by-step, exactly what you are doing.

Following the first publication of the general technique (and the
bootstrap) in 1969 by Julian Simon and subsequent independent
development by Bradley Efron, resampling has become state-of-the-art

Who Should Attend?

Statisticians and data analysts who need to test hypotheses and
ascertain the reliability of inferences they make from samples. This
includes decision-makers, consultants, quality control analysts,
industrial statisticians and biostatisticians, among others. No
specific prerequisites, though a familiarity with basic probability
and descriptive statistics will be helpful.

I. A Brief History and Roadmap

-- Early origins of resampling
-- Recent reemergence
-- Theory of resampling

II. Hypothesis testing

Resampling makes hypothesis-testing -- too often confusing --
transparently clear.

-- Measured data and counted data
-- Single sample, two samples, multiple samples
-- Bootstrap tests, permutation tests, and tests of proportions
-- Correlation
-- Regression
-- Imbalanced sample sizes

III. Confidence Intervals

-- Bootstrap confidence intervals
-- Measured data and counted data
-- Sample size determination

IV. Resampling's Relationship to Conventional Parametric Methods

Cost: $1095 (prepaid) Academic discounts available.

Institute for Professional Education
2200 Wilson Blvd., Ste. 406
Arlington, VA 22201
phone: 703-527-8700, fax: 703-527-8703
email coursesipe@aol.com


The Gordian Institute will debut two new courses designed to introduce
corporate decision makers to data mining and market timers to advanced
predictive technologies. The new seminars are designed to offer
intensive reviews of terminology, as well as benefits and pitfalls
while minimizing time away from the office. The data mining course
provides an optional third day hands-on workshop.

2. "Data Mining: Principles & Practice," January 28, 1998: Santa
Clara, CA. February 11, 1998: Satellite Beach, FL (Orlando Area)

"Data Mining: Principles and Practice" will host its initial offerings
in Santa Clara, CA in January, and Orlando, FL in February. At $995,
the two day data mining seminar covers the subject of data mining from
the ground up. Those in attendance will learn about different methods
of modeling and how those models apply to real business problems.
Those who desire to make data mining an integral part of their
business process are target candidates for Gordian's new offering.
Attendees will learn to:

     - Uncover valuable information buried in data.
     - Learn what data has real meaning and what data simply takes up
       space (also known as "data prospecting")
     - Examine which data mining methods and tools are most effective
     - Avoid pitfalls in the analysis of results

The rapid emergence of electronic data processing and collection
methods has lead some to call recent times as the "Information Age."
However, it may be more accurately termed as "The Age of the Data
Glut." Most businesses either posses a large database or have access
to one. These databases contain so much data that it becomes very
difficult to understand what that data is telling us. There is hardly
a transaction that does not generate a computer record somewhere.

All this data has meaning with respect to better understanding
customer needs and preferences. But how do you discover those needs
and preferences in a database that contains gigabits of seemingly
incomprehensible numbers and facts? Data mining does just that.
However, used blindly, incorporation of data mining techniques can
result in large expenditures of money and time to no avail. The key
issue explored in this seminar is how to avoid frustrating and costly
mistakes and improve your business process by correct use of these
powerful methods. Attendees will learn:

     - The basic principles of data mining
     - The different methods of data mining and how they compare
     - How to prepare raw data for data mining
     - How to analyze and validate the results
     - What questions data mining can answer
     - What are the pitfalls and how to avoid them
     - What commercial products are available and how to evaluate them

Gordian's "Data Mining: Principles and Practice" seminar focuses on
actual use and implementation of data mining techniques in the real
world. The instructor has been deeply involved with the development
of data mining methods and the means of their use. Actual products
will be reviewed, as will results drawn from real data mining

Those who would like a hands-on perspective to the instructional
sessions may attend an optional third day application workshop for an
additional $495. The workshops will highlight superior performance as
well as pitfalls resulting from various tools and techniques when
applied to different types of data intensive problems. Exercises will
reveal impressive results from the same technique that may have failed
in another category.

In addition, objective evaluations of popular data mining products can
save immeasurable time and effort in assessing and selecting which
suite of tools will perform best for your application. The instructor
will show how to evaluate various packages based on strengths,
limitations, value and general performance. Products will be
separated into four categories:

     - Statistical
     - Decision Tree
     - Neural Net
     - Clustering Technologies

The presenter, Ben A. Hitt, Ph.D. has many years of experience using
pattern recognition technologies and intelligent software tools to
solve business problems. He has taught thousands of students in the
use and principles of advanced software and machine learning
technologies. He was Director of Training for NeuralWare, Inc. in
Pittsburgh, PA, and in that role instructed the use of neural networks
for Financial Forecasting, Fraud Detection, Process Control and Direct
Marketing. He was instrumental in the design and development of
ModelMAX, a complete neural network application for the direct
marketing industry. Dr. Hitt also designed and implemented a
nationally recognized detection system for rapid tax refund
application fraud. He has recently conducted a detailed and
exhaustive survey of commercial data mining products for a major US

(contact information follows #3)

3. "Advanced Techniques for the Analysis of Financial Markets,"
January 20, 1998, Herndon, VA (a suburb of Washington, DC)

The one-day seminar "Advanced Techniques for the Analysis of Financial
Markets" provides attendees with a methodology for developing a
trading strategy for financial instruments. The course addresses the
selection of financial instruments to monitor, establishing
performance objectives, and techniques for the development of a
"library" of trading scenarios. This course will debut at $995 in
Washington, DC in January.

"Advanced Techniques for the Analysis of Financial Markets" is
intended for those who seek better than average performance from their
investment decision making. It is designed for the individual ready
and willing to explore approaches that have the potential to
dramatically outperform the market averages.

Recent research has proven that the financial markets are driven as
much by buyer behavior as by pure economics. This course abandons the
ideas behind capital market theory, portfolio theory and random walk.
In their place, we develop methods of identifying cases where
significant profit potential exists while screening out mediocre

The methods employed are geared toward individuals who trade financial
instruments for profit, not to "own a good company." The techniques
are most applicable to short-term and intermediate-term trading.

"Advanced Techniques for the Analysis of Financial Markets" focuses on
the development and implementation of techniques that can be directly
applied to trading financial instruments in a manner consistent with
the attendees goals and objectives. The instructor presents a
development methodology that allows attendees to identify trading
opportunities with a high probability of success.

The presenter, Thomas A. "Tony" Rathburn left his teaching and
research position at Kent State University in 1992 to co-develop the
course Applying Neural Computing to Market Timing for NeuralWare, Inc.
He is the author of numerous publications and has extensive
consulting experience in the application of advanced analysis
technologies to the financial sector. Mr. Rathburn currently provides
consulting services to a variety of organizations on the application
of advanced technologies.

Reserve your seat early, as course sizes are limited to allow for a
high level of interaction with the instructors. Additional details
for both Gordian courses, to include course outlines, specific dates,
training sites and registration information may be obtained through
any of the following:

- Email: agent@gordianknot.com
  (Send your message with any of the following as the SUBJECT)
              - Data Mining Details
              - Financial Markets Details
              - Quarterly Electronic Newsletter
- Web: http://www.gordianknot.com
- Toll Free: 800-405-2114
- Direct: 281-364-9882
- Fax: 281-754-4014


----------------------------Original message----------------------------