Aim of the Course: This
four-week course will
provide a practical guide
to multivariate analysis,
with the emphasis on the
use of the bootstrap,
decision trees, and
permutation tests in
conjunction with
parametric techniques to
analyze large arrays of
data similar to those
collected during DNA, RNA,
and protein sequencing,
EEG, MRI, fMRI, PET,
telemetry and other forms
of image analysis.
Who
Should Take This Course:
Geneticists, protein
chemists,
neurophysiologists,
epidemiologists,
geographers, astronomers,
agronomists, other image
analysts, and
statisticians.
Instructor:
Dr. Phillip Good, former
Calloway Professor of
Computer Science at the
University of Georgia
(Fort Valley) and graduate
of the program in
mathematical statistics at
UC Berkeley, is the author
of Analyzing
the Large Number of
Variables in Biomedical
and Satellite
Imagery (Wiley,
2011). Common Errors in
Statistics (and How to
Avoid Them) (Wiley,
4th ed., 2012 with James
Hardin), Permutation,
Parametric, and Bootstrap
Tests of Hypotheses
(Springer, 3rd ed, 2005), Manager's
Guide to Design and
Conduct of Clinical Trials
(Wiley, 2nd ed., 2006),
and Applying Statistics
in the Courtroom (CRC,
2001). He has given
tutorials at the Joint
Statistical Meetings
(U.S.) and Deming
Conference, lectured in
Belgium, Bulgaria, France,
Holland, Ireland,
Slovenia, and Spain, and
was twice a traveling
lecturer for the American
Statistical Association.
This is his 8th year of
providing on-line
interactive courses.
Prerequisite:
You should have
familiarity with basic
parametric statistical
concepts. Resampling
methodology will be
introduced as needed. The
relevant biomedical
background is also
provided as needed.
Organization
of the Course: The
course takes place over
the Internet. During each
course week, you
participate at times of
your own choosing - there
are no set times when you
must be online. Course
participants will be given
an alias and access to a
private bulletin board
that serves as a forum for
discussion of ideas,
problem solving, and
interaction with the
instructor.
The
course is scheduled to
take place over four
weeks, and should require
no more than ten hours per
week. At the beginning of
each week, participants
receive the relevant
material, in addition to
answers to exercises from
the previous session.
During the week,
participants are expected
to go over the course
materials and work through
exercises. Discussion
among participants is
encouraged. The instructor
will provide answers and
comments.
Textbook:
Analyzing
the Large Number of
Variables in Biomedical
and Satellite
Imagery.
Wiley, 2011.
Course
Program: The course is
structured as follows
SESSION
1: Basic Concepts
- Analyzing Very Large
Arrays of
Data--Problems and
Solutions
- Advantages of and
Necessity for
Multivariate Analysis.
- Hotelling's T
2
- Using Permutation
Tests to Establish
Statistical
Significance
- Combining
Independent Tests
- Software
SESSION
2: Testing Multiple
Hypotheses
- Reducing the Number
of Variables
- Controlling the
Over-All Error
Rate--Examples
- Controlling the
False Discovery
Rate--Time-Course Data
- Gene Set Enrichment
Analysis
- Software
SESSION 3:
Applying the Bootstrap
and Permutation Tests
- Pre-Post
Comparisons
- The Bootstrap
- Determining Sample
Size
- Validating a
Cluster Analysis
- Bootstrap or
Permutation Test?
SESSION
4: Classifying
Subjects on the Basis of
Biomedical Data
- Classification
Methods
- Decision Trees
- Misclassification
Costs
- Ensemble Methods
- Validation
- Software for Use
with Large Numbers of
Variables
