A project-oriented course in probability and statistics for undergraduate electrical engineering students

An introductory course in probability and statistics for third-year and fourth-year electrical engineering students is described. The course is centered around several computer-based projects that are designed to achieve two objectives. First, the projects illustrate the course topics and provide hands-on experience for the students. The second and equally important objective of the projects is to convey the relevance and usefulness of probability and statistics to practical problems that undergraduate students can appreciate. The benefit of this course as to motivate electrical engineering students to excel in the study of probability concepts, instead of viewing the subject as one more course requirement toward graduation. The authors co-teach the course, and MATLAB is used for mast of the computer-based projects

Accounting for Fetal Death: Vital Statistics and the Medicalization of Pregnancy in the United States

Recent reports by the Centers for Disease Control and Prevention have decried the high rate of fetal mortality in the contemporary United States. Much of the data about fetal and infant deaths, as well as other poor pregnancy outcomes, are tabulated and tracked through vital statistics. In this article, I demonstrate how notions of fetal death became increasingly tied to the surveillance of maternal bodies through the tabulating and tracking of vital statistics in the middle part of the twentieth century. Using a historical analysis of the revisions to the United States Standard Certificate of Live Birth, and the United States Standard Report of Fetal Death, I examine how the categories of analysis utilized in these documents becomes integrally linked to contemporary ideas about fetal and perinatal death, gestational age, and prematurity. While it is evident that there are relationships between maternal behavior and birth outcomes, in this article I interrogate the ways in which the surveillance of maternal bodies through vital statistics has naturalized these relationships. Copyright 2013 Elsevier Ltd. All rights reserved.

Simultaneous Multiscale Polyaffine Registration by Incorporating Deformation Statistics

Locally affine (polyaffine) image registration methods capture intersubject non-linear deformations with a low number of parameters, while providing an intuitive interpretation for clinicians. Considering the mandible bone, anatomical shape differences can be found at different scales, e.g. left or right side, teeth, etc. Classically, sequential coarse to fine registration are used to handle multiscale deformations, instead we propose a simultaneous optimization of all scales. To avoid local minima we incorporate a prior on the polyaffine transformations. This kind of groupwise registration approach is natural in a polyaffine context, if we assume one configuration of regions that describes an entire group of images, with varying transformations for each region. In this paper, we reformulate polyaffine deformations in a generative statistical model, which enables us to incorporate deformation statistics as a prior in a Bayesian setting. We find optimal transformations by optimizing the maximum a posteriori probability. We assume that the polyaffine transformations follow a normal distribution with mean and concentration matrix. Parameters of the prior are estimated from an initial coarse to fine registration. Knowing the region structure, we develop a blockwise pseudoinverse to obtain the concentration matrix. To our knowledge, we are the first to introduce simultaneous multiscale optimization through groupwise polyaffine registration. We show results on 42 mandible CT images.

Squared Extrapolation Methods (SQUAREM): A New Class of Simple and Efficient Numerical Schemes for Accelerating the Convergence of the EM Algorithm

We derive a new class of iterative schemes for accelerating the convergence of the EM algorithm, by exploiting the connection between fixed point iterations and extrapolation methods. First, we present a general formulation of one-step iterative schemes, which are obtained by cycling with the extrapolation methods. We, then square the one-step schemes to obtain the new class of methods, which we call SQUAREM. Squaring a one-step iterative scheme is simply applying it twice within each cycle of the extrapolation method. Here we focus on the first order or rank-one extrapolation methods for two reasons, (1) simplicity, and (2) computational efficiency. In particular, we study two first order extrapolation methods, the reduced rank extrapolation (RRE1) and minimal polynomial extrapolation (MPE1). The convergence of the new schemes, both one-step and squared, is non-monotonic with respect to the residual norm. The first order one-step and SQUAREM schemes are linearly convergent, like the EM algorithm but they have a faster rate of convergence. We demonstrate, through five different examples, the effectiveness of the first order SQUAREM schemes, SqRRE1 and SqMPE1, in accelerating the EM algorithm. The SQUAREM schemes are also shown to be vastly superior to their one-step counterparts, RRE1 and MPE1, in terms of computational efficiency. The proposed extrapolation schemes can fail due to the numerical problems of stagnation and near breakdown. We have developed a new hybrid iterative scheme that combines the RRE1 and MPE1 schemes in such a manner that it overcomes both stagnation and near breakdown. The squared first order hybrid scheme, SqHyb1, emerges as the iterative scheme of choice based on our numerical experiments. It combines the fast convergence of the SqMPE1, while avoiding near breakdowns, with the stability of SqRRE1, while avoiding stagnations. The SQUAREM methods can be incorporated very easily into an existing EM algorithm. They only require the basic EM step for their implementation and do not require any other auxiliary quantities such as the complete data log likelihood, and its gradient or hessian. They are an attractive option in problems with a very large number of parameters, and in problems where the statistical model is complex, the EM algorithm is slow and each EM step is computationally demanding.

Bayes Factors Based on Test Statistics

Traditionally, the use of Bayes factors has required the specification of proper prior distributions on model parameters implicit to both null and alternative hypotheses. In this paper, I describe an approach to defining Bayes factors based on modeling test statistics. Because the distributions of test statistics do not depend on unknown model parameters, this approach eliminates the subjectivity normally associated with the definition of Bayes factors. For standard test statistics, including the _2, F, t and z statistics, the values of Bayes factors that result from this approach can be simply expressed in closed form.

Impact factor statistics and publication practice: What can we learn? (Editorial)

Peer review procedures and citation statistics are important yet often neglected components of the scientific publication process. Here I discuss fundamental consequences of such quality measures for the scientific community and propose three remedial actions: (1) use of a ''Combined Impact Estimate'' as a measure of citation statistics, (2) adoption of an open reviewing policy and (3) acceleration of the publication process in order to raise the reputation of the entire discipline (in our case: behavioural science). Authors, reviewers and editors are invited to contribute to the improvement of publication practice.