On Bayes and unbiased estimators of loss

We consider estimation of loss for generalized Bayes or pseudo-Bayes estimators of a multivariate normal mean vector, θ. In 3 and higher dimensions, the MLEX is UMVUE and minimax but is inadmissible. It is dominated by the James-Stein estimator and by many others. Johnstone (1988, On inadmissibility of some unbiased estimates of loss,Statistical Decision Theory and Related Topics, IV (eds. S. S. Gupta and J. O. Berger), Vol. 1, 361–379, Springer, New York) considered the estimation of loss for the usual estimatorX and the James-Stein estimator. He found improvements over the Stein unbiased estimator of risk. In this paper, for a generalized Bayes point estimator of θ, we compare generalized Bayes estimators to unbiased estimators of loss. We find, somewhat surprisingly, that the unbiased estimator often dominates the corresponding generalized Bayes estimator of loss for priors which give minimax estimators in the original point estimation problem. In particular, we give a class of priors for which the generalized Bayes estimator of θ is admissible and minimax but for which the unbiased estimator of loss dominates the generalized Bayes estimator of loss. We also give a general inadmissibility result for a generalized Bayes estimator of loss. Research supported by NSF Grant DMS-97-04524.     

Approximation of multi-scale elliptic problems using patches of finite elements

In this paper we present a method to solve numerically elliptic problems with multi-scale data using multiple levels of not necessarily nested grids. The method consists in calculating successive corrections to the solution in patches whose discretizations are not necessarily conforming. It resembles the FAC method (see Math. Comp. 46 (174) (1986) 439–456) and its convergence is obtained by a domain decomposition technique (see Math. Comp. 57 (195) (1991) 1–21). However it is of much more flexible use in comparison to the latter. To cite this article: R. Glowinski et al., C. R. Acad. Sci. Paris, Ser. I 337 (2003).     

Evidences of phase separation in moisture‐cured poly(urethane urea)s by means of temperature modulated differential scanning calorimetry

The degree of phase separation in several moisture‐cured poly(urethane urea)s (PUUs) was studied by FTIR spectroscopy, wide angle X‐ray diffraction (WAXD), and temperature‐modulated differential scanning calorimetry (TMDSC). This latter technique was shown to be particularly useful in analysing the degree of phase separation in PUU polymers. Both phase mixing and phase segregation coexisted in the PUUs and the degree of phase separation increased as the urea hard segment (HS) content in the PUU increased. The maximum solubility of urea HSs into the polyol soft segments (SSs) was achieved for 50 wt % urea HS content in diol‐based PUUs, whereas for triol‐based PUUs the highest solubility between HS and SS was reached for lower urea HS amount. Finally, the higher the urea HS content the higher the extent of phase separation in the PUU. © 2007 Wiley Periodicals, Inc. J Polym Sci Part B: Polym Phys 45: 3034–3045, 2007     

Application of a generalized residual model to frequency domain modal parameter estimation

Residuals have long been used in frequency domain parameter estimation methods to model the influence of out-of-band modes but, typically, as a fixed set of no more than two or three terms. Recently, a systematic approach to the use of residual polynomials has led to the development of a generalized residual model. The use of a generalized residual model with rational fraction polynomial frequency domain parameter estimation methods allows the contribution of out-of-band modes to be included without increasing the model order and creating additional computational poles. Of particular interest is the use of the generalized residual for single-degree-of-freedom (s.d.o.f.) techniques, which generally do not consider the residual effects and suffer accordingly. With the use of generalized residuals, it becomes possible to properly account for nearby modes and also extract accurate residues with an s.d.o.f. algorithm. The development of the generalized residual polynomial model is outlined and a new s.d.o.f. frequency domain algorithm with generalized residuals is developed.     

Neurophysiological Model of the Normal and Abnormal Human Pupil

A model of the pathways controlling the size of the human pupil is presented. Computer simulation of this model demonstrates the role played by each of the elements in the pupil pathways. Simulations of the effects of drugs and a few common abnormalities in the system also help to illustrate the workings of the internal processes. Computer models of this type can be used as teaching aids or as tools for testing of hypotheses regarding the system.     

MRI fuzzy segmentation of brain tissue using neighborhood attraction with neural-network optimization.

Image segmentation is an indispensable process in the visualization of human tissues, particularly during clinical analysis of magnetic resonance (MR) images. Unfortunately, MR images always contain a significant amount of noise caused by operator performance, equipment, and the environment, which can lead to serious inaccuracies with segmentation. A robust segmentation technique based on an extension to the traditional fuzzy c-means (FCM) clustering algorithm is proposed in this paper. A neighborhood attraction, which is dependent on the relative location and features of neighboring pixels, is shown to improve the segmentation performance dramatically. The degree of attraction is optimized by a neural-network model. Simulated and real brain MR images with different noise levels are segmented to demonstrate the superiority of the proposed technique compared to other FCM-based methods. This segmentation method is a key component of an MR image-based classification system for brain tumors, currently being developed. Index Terms-Improved fuzzy c-means clustering (IFCM), magnetic resonance imaging (MRI), neighborhood attraction, segmentation.     

Synthesis of optically active 6-substituted 2-(aminomethyl)chromans

Three novel, optically active, 6-substituted 2-(aminomethyl)chromans were synthesized from readily available chroman 2-carboxylic acid precursors. These chroman-containing primary amines are useful building blocks for the synthesis of chroman-derived pharmaceutical agents.     

Preparation and characterization of finely dispersed drugs. 5. Ethyl 3,5-di(acetylamino)-2,4,6-triiodobenzoate

Controlled precipitation of the diagnostic imaging agent ethyl 3,5-di(acetylamino)-2,4,6-triiodobenzoate has been used to produce fine particles of various sizes, morphologies, and degrees of crystallinity, which depended on experimental conditions. In addition, two distinct polymorphic forms of the drug have been fully characterized by single crystal X-ray diffraction studies, and evidence for a third polymorph was also observed. Some of the so prepared dry particles were coated with a thin layer of silica.     

Planning District‐Wide Professional Development: Insights Gained From Teachers and Students Regarding Mathematics Teaching in a Large Urban District

This paper describes the mechanism used to gain insights into the state of the art of mathematics instruction in a large urban district in order to design meaningful professional development for the teachers in the district. Surveys of close to 2,000 elementary, middle school, and high school students were collected in order to assess the instructional practices used in mathematics classes across the district. Students were questioned about the frequency of use of various instructional practices that support the meaningful learning of mathematics. These included practices such as problem solving, use of calculators and computers, group work, homework, discussions, and projects, among others. Responses were analyzed and comparisons were drawn between elementary and middle school students' responses and between middle school and high school responses. Finally, fifth‐grade student responses were compared to those of their teachers. Student responses indicated that they had fewer inquiry‐based experiences, fewer student‐to‐student interactions, and fewer opportunities to defend their answers and justify their thinking as they moved from elementary to middle school to high school. In the elementary grades students reported an overemphasis on the use of memorization of facts and procedures and sparse use of calculators. Results were interpreted and specific directions for professional development, as reported in this paper, were drawn from these data. The paper illustrates how student surveys can inform the design of professional development experiences for the teachers in a district.     

Matrix extensions and eigenvalue completions, the generic case

In this paper we provide new necessary and sufficient conditions for the so-called eigenvalue completion problem.