-------------------------------------------09/17/2004--------------------------------------

Title: Inference on Abundance and Analysis of Spatial Patterns in Ecological Communities

Speaker: Professor Sujay Datta
Dept. of Mathematics, Statistics and Computer Science
Northern Michigan University, Marquette, Michigan, USA
Date: Sept 17th, 2004
Location: Monroe Hall 307, 2115 G St., NW
Time: 11:00am -12:00 noon.

Abstract : Ecology is the study of animal and plant populations on the face of the earth: their habitats, behavior, resources and mutual interactions (cooperation, competition, etc.). The cornerstone of many (though not all) studies in single-population ecology as well as community ecology is an estimate of the abundance of a particular population. It can be determined in an absolute or a relative sense (the latter being preferred due to its practical convenience) and by means of either complete enumeration or ‘fair and representative’ sampling (the latter being more common since a census is often not feasible or practical). The sequence of decisions by which one decides how to estimate absolute or relative abundance involves many factors: ecological, economic and statistical. In landscape ecology, on the other hand, a central issue is that of patterns of individuals in space. Considering that organisms are seldom spread across the landscape at random, this is an important question in both behavioral ecology (where zoologists are concerned with the spacing behaviors of animals) and plant ecology (where botanists often study plants as individuals).  A variety of statistical methods have been put forward over several decades by ecologists for both of these problems.
          This presentation provides a brief overview of some of these methods. Regarding abundance estimation, we touch upon the four broad categories of procedures: capture-mark-recapture techniques, removal and resight methods, methods based on quadrat counts and those based on line-transects and distances. Regarding spatial patterns, we first address the question of what pattern a population exhibits and then move on to the question of developing a set of distance measures that help us make comparisons between patterns in two populations. Methods discussed include various tests for spatial patterns and various indices of dispersion appropriate for different scenarios (e.g., when a complete spatial map is available, when such a map is unavailable and sampling is needed, when sampling units are natural or arbitrary). Time permitting, other sampling-related issues (such as adaptive sampling, multistage/sequential sampling) are discussed.

-------------------------------------------09/24/2004--------------------------------------

Title: Symposium On Frontiers of Statistical, Mathematical and Computational Sciences

Location: The Marvin Center, George Washington University , 800 21st Street NW , Washington , DC
Date: Sept 24th, 2004
Time: 08:30am -04:00 pm.

Agenda:

• 0845-0900 Opening Remarks/Announcements
• Session 1 Chair: David C. Arney, Army Research Office
• 0900-1000 Statistical Models for Simulation Errors and their Role in Prediction and UncertaintyQuantification, Professor James Glimm , SUNY at Stony Brook (Introduced by Donald Lehman, Executive VP for Academic Affairs, GWU)
• 1000-1030 Break
• 1030-1130 Embracing Statistical Challenges in the Information Technology Age, Professor Bin Yu, University of California at Berkeley (Introduced by Efstathia Bura, GWU )
• 1130- 1300 Lunch Break
• Session 2 Chair: Mou-Hsiung Chang, Army Research Office
• 1300-1400 Extreme Events and Large Deviation Theory, Professor S. R. S. Varadhan, Courant Institute of Mathematical Sciences (Introduced by Tapan Nayak, GWU)
• 1400-1500 ComputerVision: A Nexus of Mathematics, Statistics and Computations, Professor David Mumford, Brown University (Introduced by Nozer Singpurwalla, GWU)
• 1500- 1600 Reception

-------------------------------------------10/08/2004--------------------------------------

Title: From Reliability to Finance

Speaker: Dr. Alex Kreinin
Principal Mathematician, Quantitative Research Branch
Algorithmics Inc., Toronto, Canada
Date: 10/08/2004
Location: Monroe Hall 307, 2115 G St., NW
Time: 11:00-12:00 noon.

Abstract : In this talk we consider an analogy between modeling of reliability of complex systems and valuation of financial credit-risky securities, usually called credit derivatives. We discuss a valuation approach to pricing of a class of these securities, its performance and also discuss interpretation of the results in Reliability Theory. This talk is based on the paper with Ian Iscoe.

-------------------------------------------10/15/2004--------------------------------------

Title: A Partial Correlation Characterization for Dirichlet-Type Distributions

Speaker: Professor T. A. Mazzuchi
 Department of Engineering Management and Systems Engineering
George Washington University
Date: 10/15/2004
Location: Monroe Hall 103, 2115 G St., NW
Time: 4:00 pm - 5:00 pm.

Abstract : The Dirichlet distribution is a multivariate distribution with many nice properties. Used primarily as a conjugate prior distribution for the multinomial distribution parameters, many of its conditional distribution properties have not been explored or utilized. In this talk, we present a characterization of the Dirichlet distribution based on its conditional distribution properties. That is, if the Dirichlet variables are partitioned in to two groups, the partial correlation for any pair in the first group is equivalent to the conditional correlation of the pair (conditioned on the second group). This is a somewhat unique distributional result. If time permits, the above result will also be demonstrated for the Ordered Dirichlet distribution as well.

-------------------------------------------10/29/2004--------------------------------------

Title: Empirical Process Approach to Some Two-Sample Problems Based on Ranked Set Samples

Speaker: Professor Kaushik Ghosh
Department of Statistics, George Washington University
Date: 10/29/2004
Location: Monroe Hall, 103, 2115 G St., NW
Time: 4:00 pm - 5:00 pm.

Abstract : For any two distribution functions F and G, the horizontal and vertical shift functions are related to the Q-Q and P-P plots, which are widely used graphical tools in assessing goodness-of-fit or in testing equality of two distributions. In this article, we study the asymptotic properties of these shift functions based on independent ranked set samples drawn from continuous distributions. As the distributions of the limiting processes depend on the unknown populations, the bootstrap is used to construct confidence bands for these shift functions. Departures from the null hypothesis F = G are detected by checking whether the confidence band completely contains the zero line. It is shown that by using balanced ranked set samples with bigger set sizes, one can decrease the width of the confidence band and hence increase the power of detection of shift. These theoretical findings are validated through simulation studies and an application to cancer mortality data.

-------------------------------------------11/05/2004--------------------------------------

Title: The Design of Computer Experiments to Determine Optimum and Robust Control Variables

Speaker: Professor William Notz
Department of Statistics, Ohio State University
Date: 11/05/2004
Location: Monroe Hall, 307, 2115 G St., NW
Time: 11:00-12:00 noon.

Abstract : In this talk I will discuss the design of computer experiments when there are two types of inputs: control variables and noise variables. Control variables are determined by a product designer while noise variables are uncontrolled in the field but take on values according to some probability distribution. I will consider two problems. The first is the situation in which there are two outputs (responses), each of which is expensive or time consuming to compute. The objective is to find values of the control variables that optimize the mean (over the distribution of the noise variables) of one response subject to a constraint on the mean of the other response. The second is to find values of the control variables at which the response is insensitive to the value of the noise variables.

For both problems, I will describe a sequential strategy to select the values of the inputs at which to observe the responses. The methodology is Bayesian; the prior takes the responses as draws from a Gaussian stochastic process. At each stage, the strategy determines which response to observe and at what set of inputs so as to maximize a posterior expected "improvement" over the current estimate of the optimum. This is joint work with Jeffrey Lehman, Tom Santner, and Brian Williams.

-------------------------------------------11/12/2004--------------------------------------

Title: Sequential Classification on Lattices with Experiment-Specific Response  Distributions, with Applications

Speaker: Professor Curtis Tatsuoka
Department of Statistics, George Washington University
Date: 11/12/2004
Location: Monroe Hall 307, 2115 G St., NW
Time: 11:00-12:00 noon.

Abstract : A statistical framework will be described for the problem when there exists a “true” state among a collection of states, and observations from sequentially selected experiments are used to identify it.  The classification model is assumed to be a lattice, and response distributions will be experiment-specific.  This generalizes a framework described by Tatsuoka and Ferguson (2003), which assumes that all experiments share the same response distributions.  Applications of this setting include those in group testing, and in neuropsychological and educational assessment.  Results relating to optimal rates of convergence will be discussed, and a simple and intuitive class of experiment selection rules will be shown to attain optimal rates under general conditions.  An application in neuropsychological assessment will be presented.

Title: Regression models with increasing number of unknown parameters

Speaker: Dr. Asaf Hajiyev
Azerbaijan Academy of Sciences
Department of Probability and Statistics, Baku State University.
Date: 11/12/2004
Location: Monroe Hall, room 103
Time: 4:00 pm - 5:00 pm.

Abstract : The regression models (linear and nonlinear) with increasing numbers of unknown parameters and unknown variances of the errors are considered. At the each point of observation there is only one observable value and that does not allow for estimation of a variance. Such problems are typical for applications, but  there haven't been enough investigations on models with increasing numbers of unknown parameters and unknown variances. The method of direct estimation (without estimation of the variances) of
the elements of the covariance matrix of deviation vector is suggested. Using this method a confidence band for unknown function in regression models has been constructed.

-------------------------------------------11/19/2004--------------------------------------

Title: Nonparametric, Hypothesis-based Analysis of Molecular Heterogeneity for Comparative Phenotype Characterization

Speaker: Professor Jeanne Kowalski
Division of Biostatistics
Johns Hopkins University
Date: November 19, 2004 – 4:00pm-5:00pm,
Location: Monroe Hall 103, 2115 G St., NW
Time: 4:00 pm - 5:00 pm.

Abstract : In this talk, I describe two novel, inference-based approaches to analysis of molecular heterogeneity associated with phenotypes.  A common theme among them is the construction of testable hypotheses in a very high-dimensional setting, based on developed U-statistic theory, with nonparametric inference.  With a modest sample, I discuss a distance-based approach for analysis of sequence heterogeneity.  In the extreme case of several single, high-dimensional samples that are to be compared from a microarray experiment, I introduce a class of stochastic linear hypotheses that includes the Mann-Whitney Wilcoxon rank sum test as a special case.  In each setting, I discuss the statistical and bioinformatic approaches developed to characterize either genes within a genome or locations within a sequence that depict groups of similar phenotype.  As motivation, I examine two separate problems, one for relating sequence heterogeneity in a region of the HIV genome to drug resistance, and a second for relating gene expressions to hypothesized pathways for immunogenetic analysis of T cells.

-------------------------------------------12/--/2004--------------------------------------

Title:

Speaker:
Department of Statistics,
Date:
Location:
Time: 11:00-12:00 noon.

Abstract :

--------------------------------------------------------------------------------
The contact person is Reza Modarres at Reza@gwu.edu or 202-994-6359.