Please check back for updates.
Each time you visit click the reload button in your browser to load updates.

SCHEDULE

ABSTRACTS

Ancient Human Social Dynamics: Object Detection in High-Resolution Satellite Imagery - Searching for Needles in Haystacks
Prem Goel
Department of Statistics
The Ohio State University
Cloumbus, OH

The AHSD Project explores the long-term development of tribal social identities in Southern Arabia through spatial analysis of the remnant markers for ancient tribal territorial behaviors. The team combines multidisciplinary expertise in archaeology, high-resolution GPS, and analysis of high resolution satellite imagery in developing new protocols for recognizing and mapping small scale stone monuments distributed across the remote desert margins of southern Yemen and Oman Archaeologists recognize that almost all of Southern Arabia's peoples before history were pastoralists and lived highly mobile lives. Rather than build houses and settlements, these groups dedicated their labor to building small stone monuments commemorative of burials and graves, sacrifices and feasts, and tribal gatherings. A team of archaeologists working on the ground has generated a good chronology and typology of monuments as the basis for examining their spatial distribution. The logistics of field studies in remote Hadramawt region of Yemen and Dhofar region of Oman make the task of visiting and ascertaining the location of each monument prohibitive; so the team has turned to modern remote sensing, imaging, mapping and statistical tools to develop automated detection of monuments from high-resolution satellite images to generate spatial and chronological distributions of monument locations, that will help reconstruct tribal territories and dynamics in territorial signaling through antiquity.
 
We are developing a statistical procedure for detecting specific types of objects, called High Circular Tombs (HCTs), from high-resolution satellite imagery. The key feature of this problem is the large variability in the background across different segments in an image and between images across the region of interest, mainly due to many landforms on which these monuments have been erected. Since each high resolution image covers a large area to be searched for monument detection, we employ statistical approach, instead of a template matching approach. Automatic monument detection research has thus explored using panchromatic intensity, shape, size, multispectral RGB and IR bands for hue, saturation and vegetation index, as well as landform-elevation as features for detecting cairns, using some of the 2008 HCT locations on imagery as a training set. The goal is to obtain a computer generated indicator that correctly identifies almost all known locations of HCT while minimizing or eliminating false positives in areas where all HCT locations are known. The details of this algorithm will be presented along with the underlying intuition.

Multifractal Fractional Sum--Difference Models for Internet Traffic
David Anderson
Mathematics Department
Xavier University of Louisiana
New Orleans, Louisiana

Internet quality of service is strongly effected by packet queuing behavior, which in turn depends on the behavior of packet interarrival times. Thus, accurate modeling of packet interarrival times is essential for maintaining a high quality of service on the Internet. Packet interarrival times have been studied before, and some models have been put forward. However, the Fractional Sum--Difference model is unique in that it offers the first parsimonious model that is able to account for statistical phenomena and provide formulas for and insight into the behavior of Internet traffic packet interarrival times. This includes the behavior of multifractal plots, variance--time (VT) plots, and the autocorrelation function (acf), as well as time--scaling behavior.
 
The simplicity of the FSD model comes from its use of a non--linear, monotone transformation of the interarrival times to a simple Gaussian process. Specifically, the Gaussian process is a long--range dependent time series that is the sum of a component which is near fractional Brownian motion (fBm) and a Gaussian white noise component. The transformation can be approximated by a logarithmic transformation, so that the untransformed interarrival times can be approximated as the product of the exponent of a near fBm component and the exponent of Gaussian white noise. The untransformed time series can be modeled as marginally Weibull, so that the FSD model has only four parameters: the mean of the untransformed interarrival times, a Weibull shape parameter for the untransformed interarrival times, a parameter indicating the relative size of the long--range dependent component of the Gaussian image, and a parameter measuring the strength of the long--range dependence (d, the Hurst parameter minus 0.5). The FSD model is highly accurate and has been validated by real--world data sets.

Variations on the Varying Coefficient Model
Brian D. Marx
Department of Experimental Statistics
Louisiana State University
Baton Rouge, LA

Although the literature on varying coefficient models (VCMs) is vast, we believe that there remains room to make these models more widely accessible and provide a unified and practical implementation for a variety of complex data settings. The adaptive nature and strength of P-spline VCMs allow for a full range of models: from simple to additive structures, from standard to generalized linear models, from one-dimensional coefficient curves to two-dimensional (or higher) coefficient surfaces, among others, including bilinear models and signal regression. As P-spline VCMs are grounded in classical or generalized (penalized) regression, fitting is swift and desirable diagnostics are available. We will see that in higher dimensions, tractability is only ensured if efficient array regression approaches are implemented. We also motivate our approaches through several examples to highlight the breadth and utility of our approach.

Acceptance Sampling Plans from Truncated Life Tests Based on the Birnbaum-Saunders Distribution for Percentiles
Yuh-Long Lio
Department of Mathematical Sciences
University of South Dakota

Time to failure due to fatigue is one of the common quality characteristics in material engineering applications. The Birnbaum-Saunders distribution has been proved to provide a better fitting for the fatigue data set than the Weibull distribution does. In this talk, the developed acceptance sampling plans for the Birnbaum-Saunders percentiles when the life test is truncated at a pre-specified time will be presented. The R package named spbsq will be used to implement the developed sampling plans with two real data sets.

Random Selection vs. Position Preference
Nola McDaniel
Department of Mathematics, Computer Science, and Statistics
McNeese State University
Lake Charles, LA

Is random selection truly random? This talk examines the issue of random selection from a small group of objects. The difference between number preference and position preference is examined and the application of the results to hypothesis testing is discussed.

Bootstrapping Possibly Misspecified Models
Mihai C. Giurcanu
Mathematics Department
University of Louisiana at Lafayette
Lafayette, LA

In this talk I will describe the asymptotic properties of various bootstrap procedures for possibly misspecified moment condition models. The main finding is that although some bootstrap procedures are consistent when the model is correctly specified, they are inconsistent when the model is incorrectly specified and vice-versa. An empirical study shows the finite sample behaviour of various resampling procedures for a possibly misspecified panel data model.