From ceginfo@u.washington.edu Tue Sep 30 08:54:52 2003
Received: from mxu3.u.washington.edu (mxu3.u.washington.edu [140.142.32.133])
	by lists.u.washington.edu (8.12.9+UW03.09/8.12.9+UW03.09) with ESMTP id h8UFsphh025836
	for <cegrads@reallists.u.washington.edu>; Tue, 30 Sep 2003 08:54:51 -0700
Received: from mxout4.cac.washington.edu (mxout4.cac.washington.edu [140.142.33.19])
	by mxu3.u.washington.edu (8.12.9+UW03.09/8.12.9+UW03.08) with ESMTP id h8UFsopm007934
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NO)
	for <cegrads@u.washington.edu>; Tue, 30 Sep 2003 08:54:51 -0700
Received: from homer18.u.washington.edu (homer18.u.washington.edu [140.142.8.18])
	by mxout4.cac.washington.edu (8.12.9+UW03.09/8.12.9+UW03.08) with ESMTP id h8UFsnQk009216
	for <cegrads@u.washington.edu>; Tue, 30 Sep 2003 08:54:50 -0700
Received: from localhost (ceginfo@localhost)
	by homer18.u.washington.edu (8.12.9+UW03.09/8.12.9+UW03.09) with ESMTP id h8UFsnQT072244
	for <cegrads@u>; Tue, 30 Sep 2003 08:54:49 -0700
Date: Tue, 30 Sep 2003 08:54:49 -0700 (PDT)
From: Civil and Environmental Engineering <ceginfo@u.washington.edu>
To: cegrads@u.washington.edu
Subject: New Course Offering
Message-ID: <Pine.A41.4.44.0309300852110.85344-100000@homer18.u.washington.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Uwash-Spam: Gauge=IIIIIIII, Probability=8%, Report='LINES_OF_YELLING, __CT, __CT_TEXT_PLAIN, __HAS_MSGID, __MIME_VERSION, __SANE_MSGID'


-------------------------------------------------------------------------

INTRODUCTION TO BAYESIAN STATISTICS STAT 498

Instructor: Thomas Richardson (tsr@stat.washington.edu)
Web-page: http://www.stat.washington.edu/tsr/s498/
Class time: MWF 1.30-2.20
Classroom: EE1 031

This course provides an introduction to the Bayesian approach to
probability and statistical inference. Bayesian methods differ from
traditional statistical methods in that they place a probability
distribution over the parameter space. This distribution may be regarded
as representing our uncertainty about the parameters before we observe any
data.

The course will begin by presenting some of the foundational justification
for this approach, such as the `Dutch' book argument. We will then
consider Bayesian inference for simple models such as the normal and
binomial. In the last part of the class we will look at some of the recent
computational advances that allow these methods to be used in practice.

-------------------------------------------------------------------------


.