|
Education and Training: Introduction to Bayesian Analysis for Scientists and Engineers |
|
Time and Location |
Introduction to Bayesian Analysis for Scientists and
Engineers
Blaza Toman Statistical Engineering Division, NIST Teusday, Thursday, Teusday February 12, 14, 19, 2002 1:00pm - 5:00pm Adminstration Building, Lecture Room E NIST Gaithersburg, MD |
Abstract |
The Bayesian approach to statistical analysis dates back
to the 1700's and has flourished in theoretical development,
but has not been widely used in practice, despite many
potential advantages, until quite recently. The great
advances in computing power of the last 10-15 years, with
parallel advances in simulation (Markov Chain Monte Carlo),
now enable experimenters and more casual users of statistics
to take advantage of Bayesian methods. The most immediately
relevant advantage to NIST scientists is the relative
ease of incorporating expert opinion and Type B uncertainties
into the statistical analysis of experimental data.
This course will provide a brief introduction to Bayesian methods by applying them to some of the specific problems often encountered by NIST scientists. These include analysis of interlaboratory expriments, including key comparisons, and some aspects of experimental design. Real examples will be used to motivate the methodology. The course will introduce students to performing such analyses using the freely available software called BUGS. Notes and a text accompany the course. Course prerequisites are rudimentary knowledge of elementary statistics and probability. |
Comments on Course |
CLASS SIZE IS LIMITED TO 40.
REGISTRATION FEE IS $150. A set of notes will be provided for the class, and a textbook (Donald Barry, Statistics A Bayesian Perspective) will be utilized. |
Further Information |
For further information, contact
|
Date created: 2/1/2002 |
[ SED Home | Education Home | Previous ] Next ] |