| Date |
Title and Speaker |
| |
April 2005 |
4/26/05
3:35 PM |
Smoothly Truncated Stable Distributions, GARCH�Models, and Option
Pricing
Christian Menn, Cornell
Abstract
(PDF format) |
4/19/05
3:35 PM |
Modeling Interference in Wireless Networks with Stable Distributions
Michael Souryal, NIST
Abstract
(PDF format) |
4/5/05
3:35 PM |
Metamathematics
Kent Miller, American University
Abstract
(PDF format) |
| |
March 2005 |
3/29/05
3:35 PM |
Finding Structure in Data Streams
William Szewczyk, National Security Agency
Abstract
(PDF format) |
3/22/05
3:35 PM |
Statistics Before Your Eyes
Bob Jernigan, American University
Abstract
(PDF format) |
3/1/05
3:35 PM |
Taking Products of Particles: Poincar� Symmetry and Representation
Towers
Nate Harshman, American University Physics Department
Abstract
(PDF format) |
| |
February 2005 |
2/22/05
3:35 PM |
Summer Research Showcase
Come find out about opportunities for students to do research at AU this
summer - and get paid! |
2/17/05
10:15 AM |
Tests for Bimodality (Note special time and location: Gray Hall
104)
Eduartas Valaitis, Yale University
When a frequency distribution exhibits multimodality, there is a lack
of universally accepted hypothesis tests that can be used to evaluate
whether the modes are due to chance. A bimodality criterion is proposed
and serves as a basis for partitioning a plane into a region of bimodality
and a region of unimodality. No unbiased test exists for this bimodality
boundary shape. Nevertheless, two synthetic tests for detecting bimodality
in a mixture of two normal components with equal variances are presented.
These proposed tests are less biased than the maximum likelihood ratio
test. |
2/15/05
3:35 PM |
Meta-Analytic Approaches in Statistical Genomics
John Stevens, Purdue University
It is becoming increasingly common for multiple laboratories to use microarray
technology to study the genetic basis of the same disease or condition
of interest in the same organism. With this technology, a laboratory can
seek to identify which genes are differentially expressed between conditions.
Differences in experimental results can arise from chance variation and
fundamental differences between experiments. Estimates of each gene's
magnitude of differential expression from multiple experiments in the
same laboratory may not be independent. In order to effectively combine
results from different laboratories to achieve a clearer understanding
of each gene's true relationship to the condition of interest, it is necessary
to account for these differences and dependencies. A meta-analytic approach
to combine results from the Affymetrix platform is developed, focusing
on the use of covariate and covariance information. Fixed effects, random
effects, and hierarchical Bayes frameworks are presented. A minimum variance
threshold approach is proposed to blend the notions of biological relevance
and statistical significance. The traditional univariate Affymetrix approach
to quantifying differential expression via the signal log ratio is extended
to the multivariate case to allow for covariance estimation. This new
approach is demonstrated using both data from a simple simulation model
and experimental data from a mouse model for multiple sclerosis. The results
of this approach are compared with those of other alternative and previously
proposed approaches to combine results from multiple microarray experiments.
|
2/10/05
10:15 AM |
Exploring goodness-of-fit and spatial correlation using components
of Tango's index of spatial clustering (Note special time and location:
Gray Hall 104)
Monica Jackson, Emory University
The ability to detect anomalies as clustering in data sets plays an important
role in spatial data analysis. Tango (1995) developed a statistic that
can be used to detect clusters in data sets. Rogerson (1999) observed
that Tango's index may be decomposed into the summation of two distinct
statistics, the first part a test of goodness-of-fit (GOF), and the second
part an index of spatial autocorrelation (SA) similar to Moran's I. In
this talk we investigate the effectiveness of Rogerson's expression of
Tango's statistic in separating GOF from SA in data sets containing clusters.
We simulate data under the null hypothesis of no clustering as well as
two alternative hypotheses. The first alternative hypothesis induces a
poor fit from the null hypothesis while maintaining independent observations
and the second alternative hypothesis induces spatial autocorrelation
while maintaining fit. Using Rogerson's decomposition and leukemia incidence
data from New York, we show graphically one is unable to statistically
distinguish poor fit from autocorrelation. |
2/9/05
1:30 PM |
Nonstandard Analysis: Calculus without ε and δ (Note
special time)
Jeff Adler, The University of Akron
Abstract
(PDF format) |
2/1/05
3:35 PM |
Artin L-functions and the Dedekind Conjecture. (Note special location:
Gray 104)
Josh Lansky, American University
Abstract
(PDF format) |
| |
January 2005 |
1/25/05
3:45 PM |
A Tour: Part A. Shaping a Trigonometric Curve to Hear a Sound
Part B. Separation of Sounds: Voice Prints, Methods of Independent Component
Analysis. (Note special location: McKinley 108)
I Lok Chang, American University
Abstract
(PDF format) |
| |
November 2004 |
11/09/04
3:35 PM |
Automorphism groups of curves and codes
Amy Ksir, US Naval Academy
Abstract : There is a type of code called an AG code that can be built
out of the points and functions on an algebraic curve. If the curve has
a finite group of symmetries, these will induce a group of permutations
of the code words. I will discuss the structure of this group representation,
and a result on when permutations of the code can induce symmetries of
the curve itself. |
| |
October 2004 |
10/26/04
3:35 PM |
The Absolute Galois Group and Dessins d'Enfants
Melanie Wood, Princeton University
The absolute Galois group of the algebraic numbers over the rationals
is one of the most important yet one of the most mysterious objects in
number theory--so mysterious that we can only even write down two of its
infinitely many elements. Grothendieck proposed a method of studying the
absolute Galois group by considering its action on certain maps between
algebraic curves. It turns out that these maps can be described completely
combinatorially, and absolute Galois group acts faithfully on a set of
objects so simple that Grothendieck called them "children's drawings."
These combinatorial objects allow us to say some concrete things about
the absolute Galois group and this action, and they also provide a simple
way of seeing how much we still don't know about the absolute Galois group.
|
10/26/04
8 PM |
Math
and Creativity - a special evening presentation by Ms.
Woods for the general public. Note special location: Ward
Building, Room 1. |
| 10/19/04 |
Charles Davies and the Mathematics Education Business
Amy Ackerberg-Hastings
Abstract
(PDF format) |
| |
September 2004 |
| 9/14/04 |
Quantum Dots
Andrew Shabaev, U. S. Naval Research Labs
Abstract
(PDF format) |
