| Date (most recent first) | Title and Speaker |
| April 2004 | |
| 4/20/04 | Shapley/Owen Analysis of the Power of States in the Electoral College Jason Richwine, American University Abstract: The paper ranks the power of individual states by creating a model that generates reasonable orderings of states (from most Democratic to least Democratic) in hypothetical elections. While California remains the most powerful state, a number of smaller states such as Wisconsin and Minnesota have emerged as pivotal players since 1980. Jason Richwine is an undergraduate student at AU double majoring in Math and Political Science. He was awarded a CAS Research Award for this project. |
| March 2004 | |
| 3/30/04 | Comparison of Estimators in Linear Regression Models with Auto-Correlated Disturbances: When is OLS Efficient? Samir Safi, American University Abstract: It is well known that the ordinary least squares (OLS) estimates in the regression model are efficient when the errors have mean zero, constant variance and are uncorrelated. In problems concerning time series, it is often the case that the errors are, in fact, correlated. It is known that OLS may not be optimal in this context. We have derived the relative efficiency of the generalized least squares (GLS) estimator to that of OLS estimator for linear and quadratic design vectors. We proved that the relative efficiency of the variance of GLS to that of OLS is invariant to scaling and shifting of the design vectors. Additionally, using computer simulations, we consider the robustness of various estimators, including estimated generalized least squares. We found that if the error structure is autoregressive, and the dependent variable is nonstochastic and linear or quadratic, the OLS performs nearly as well as its competitors. For other forms of the dependent variable, we have developed rules of thumb to guide practitioners in their choice of estimators |
| 3/23/04 | Stability of Hybrid Dynamic Systems Modulated by Singularly Perturbed Markov Chains
Grazyna Badowski Abstract: This work is concerned with stability of hybrid dynamic systems involving singularly perturbed Markov processes. Due to the singular perturbation and the complexity of the system, the stability analysis is difficult. Using the limit dynamic systems of the singularly perturbed processes and perturbed Liapunov function methods, the asymptotic behavior of the dynamic systems are examined. It is shown that if the limit systems are stable, then the original systems are also stable. |
| 3/2/04 | Special values of Zeta Functions Niranjan Ramachandran, University of Maryland College Park This is an expository talk on what these functions are, and what is known and conjectured about their special values. |
| 12/02/03 |
Geometric structures on surfaces and the fundamental group
William Goldman, University of Maryland |
| December 2003 | |
| 12/02/03 | The History of Algebra and Its Application to Teaching
Victor Katz, University of the District of Columbia |
| November 2003 | |
| 11/18/03 | Simple Models of Card Shuffling
Robert Jernigan, American University This talk will be accessible to undergraduate students. Free pizza and refreshments after the talk! |
| October 2003 | |
| 10/28/03 | The Arithmetic-Geometric Mean, Complete Elliptic Integrals and the Pendulum Michael Pearson, MAA |
| 10/21/03 | Stable distributions, experimental mathematics, and reversing the triangle inequality John Nolan, American University |
| September 2003 | |
| 9/23/03 | Where topology, geometry, and dynamical systems meet Richard Brown, American University |
| 9/16/03 | Ordinal Data Inference Gideon Samid President & CEO, ClearBIT Systems: A Division of AGS Encryptions Ltd. |
