Department of Statistics Seminar
North Carolina State University
presents
Dr. Montserrat Fuentes, Dr. Sharon Lubkin & Dr. Spencer Muse
North Carolina State University
A Sampler of NCSU Faculty
ABSTRACT
Title: Testing for separability of spatial-temporal covariance functions
Abstract: Classical geostatistics and Fourier spectral methods are powerful tools to study the spatial temporal structure of stationary processes with separable covariance functions. However, it is widely recognized that in real applications spatial temporal processes are rarely stationary and separable; then an important extension of these spectral methods is to processes that are nonstationary and nonseparable. In this work, some new spectral approaches and tools to test for nonstationarity and separability are presented. All the work that has been done so far to test the non-stationarity of a random process is for one-dimensional processes primarily. The approach proposed here is based on a spectral representation of a spatial temporal process, and the proposed method consists essentially in testing the homogeneity of a set of spatial spectra evaluated at different locations in space and time. Novel nonparametric approaches are presented to estimate the spatial temporal structure of a nonstationary and nonseparable spatial process defined on a continuous space. The methods are based on a spectral approach, using spectral functions that are space dependent.
Applications include modeling and testing for non-stationary and separability of air pollution fluxes and concentrations over different geo-political boundaries.
Title: Multiphase fluid mechanics in tissue dynamics
Abstract: Tissues that are growing and changing shape, whether in development, maintenance, or disease, are deformed by forces which they generate themselves. Ongoing work in our group aims to understand how tissues can deform themselves. The approach involves a continuum approximation of tissue properties, and results in multiphase fluid models.
Title: Studying the Evolution of Protein-Coding DNA Sequences
Abstract: A fundamental goal in the analysis of DNA sequence data is to understand how rates of DNA sequence evolution vary among organisms, among genes, and among regions within genes. When studying DNA sequences that encode proteins, there are two major types of sequence changes: nonsynonymous changes that result in a change in the encoded protein, and synonymous changes that have no affect on the protein. I will discuss stochastic models for describing the evolution of these sequences and inference procedures for identifying various types of rate heterogeneities.
Friday, January 10, 2003
3:35 - 4:35 pm
206 Cox Hall
Refreshments will be served on the second floor of Dabney Hall (left of Room 222) at 3:00 pm.