MATHEMATICS DEPARTMENT
SCHEDULE
| 9:30-10:00 (MDD 451) | Reception/Coffee |
| 10:00-11:30 (MDD 209) |
The AVE method for variance component estimation R. R. Hocking Texas A & M University |
| 11:30-1:00 | Lunch |
| 1:00-1:20 (MDD 208) | Business meeting |
| 1:30-3:00 (MDD 208) |
I-Projections onto Intersections of Convex Sets Bhaskar Bhattacharya Southern Illinois University |
| 3:00-3:15 (MDD 208) | Coffee break |
| 3:15-4:15 (MDD 208) |
A Computational Approach to Statistical Inferences Nabendu Pal University of Louisiana at Lafayette |
ABSTRACTS
The AVE method for variance component estimation
R. R. Hocking
Texas A & M University
I will describe a method, called AVE, for estimating variance components in balanced, mixed linear models. The method is motivated by a formulation of the model based on the implied covariance structure of the data. An alternative computational procedure is suggested but of more importance is the diagnostic information that is provided. This information allows us to assess the data and other features of the model. The method will be demonstrated with several numerical and conceptual examples.
I-Projections onto Intersections of Convex Sets
Bhaskar Bhattacharya
Southern Illinois University
(IPFP) has been introduced formally by Deming and Stephan in 1940. For bivariate densities, this procedure has been investigated by Kullback (1968) and Ruschendorf (1995). It is well-known that the IPFP is a sequence of successive I-projections onto sets of probability measures with fixed marginals. However, when finding the I-projection onto the intersection of arbitrary closed, convex sets (e.g. marginal stochastic orders), a sequence of successive I-projections onto those sets may not lead to the actual solution. Addressing this situation, we present a new iterative I-projection algorithm. Under reasonable assumptions and using tools from Frenchel duality, convergence of this algorithm to the true solution is shown. The cases of infinite dimensional IPFP and marginal stochastic orders are worked out in this context.
A Computational Approach to Statistical Inferences
Nabendu Pal
University of Louisiana at Lafayette
The purpose of this paper is to provide a step by step computational approach to handle statistical inferences based on a parametric model for a given data set. This approach may come handy in those cases where the sampling distributions are not easy to derive or extremely complicated. Our suggested approach can be implemented mechanically by applied researchers to draw statistical inferences when a suitable parametric model is assumed for a given data set.