Model-based clustering with sparse covariance matrices

Statistics and Computing - Finite Gaussian mixture models are widely used for model-based clustering of continuous data. Nevertheless, since the number of model parameters scales quadratically with...     

Conditional covariance penalties for mixed models

The prediction error for mixed models can have a conditional or a marginal perspective depending on the research focus. We introduce a novel conditional version of the optimism theorem for mixed models linking the conditional prediction error to covariance penalties for mixed models. Different possibilities for estimating these conditional covariance penalties are introduced. These are bootstrap methods, cross-validation, and a direct approach called Steinian. The behavior of the different estimation techniques is assessed in a simulation study for the binomial-, the t-, and the gamma distribution and for different kinds of prediction error. Furthermore, the impact of the estimation techniques on the prediction error is discussed based on an application to undernutrition in Zambia.     

Building a Human Health Risk Assessment Ontology (RsO): A Proposed Framework

Over the last decade the health and environmental research communities have made significant progress in collecting and improving access to genomic, toxicology, exposure, health, and disease data useful to health risk assessment. One of the barriers to applying these growing volumes of information in fields such as risk assessment is the lack of informatics tools to organize, curate, and evaluate thousands of journal publications and hundreds of databases to provide new insights on relationships among exposure, hazard, and disease burden. Many fields are developing ontologies as a way of organizing and analyzing large amounts of complex information from multiple scientific disciplines. Ontologies include a vocabulary of terms and concepts with defined logical relationships to each other. Building from the recently published exposure ontology and other relevant health and environmental ontologies, this article proposes an ontology for health risk assessment (RsO) that provides a structural framework for organizing risk assessment information and methods. The RsO is anchored by eight major concepts that were either identified by exploratory curations of the risk literature or the exposure‐ontology working group as key for describing the risk assessment domain. These concepts are: (1) stressor, (2) receptor, (3) outcome, (4) exposure event, (5) dose‐response approach, (6) dose‐response metric, (7) uncertainty, and (8) measure of risk. We illustrate the utility of these concepts for the RsO with example curations of published risk assessments for ionizing radiation, arsenic in drinking water, and persistent pollutants in salmon.     

Comparing the U.S. Decennial Census Coverage Estimates for Children from Demographic Analysis and Coverage Measurement Surveys

Following every U.S. decennial census since 1960, the U.S. Census Bureau has evaluated the completeness of coverage using two different methods. Demographic analysis (DA) compares the census counts to a set of independent population estimates to infer coverage differences by age, sex, and race. The survey-based approach (also called dual system estimation or DSE) provides coverage estimates based on matching data from a post-enumeration survey to census records. This paper reviews the fundamentals of the two methodological approaches and then initially examines the results of these two methods for the 2010 decennial census in terms of consistency and inconsistency for age groups. The authors find that the two methods produce relatively consistent results for all age groups, except for young children. Consequently, the paper focuses on the results for children. Results of the 1990, 2000, and 2010 decennial censuses are shown for the overall population in this age group and by demographic detail (age, race, and Hispanic origin). Among children, the DA and DSE results are most inconsistent for the population aged 0–4 and most consistent for ages 10–17. Results also show that DA and DSE are more consistent for Black than non-Black populations. The authors discuss possible explanations for the differences in the two methods for young children and conclude that the DSE approach may underestimate the net undercount of young children due to correlation bias.     

Sixty years of change in Tibetan fertility: An assessment

Existing knowledge of Tibetan historical population development is mostly based on ‘best-guess’ estimates and is heavily politicized. Using census data, I reconstruct the development of Tibetan fertility in China since the 1940s, with the objective of providing an independent assessment that can be used as benchmark for future studies and debates on Tibetan demography. Following major social and economic transformations starting in the 1950s, Tibetan fertility unexpectedly increased from the late 1950s to the late 1960s. As noted in several existing studies, Tibetan fertility in China then declined swiftly from the early 1980s onwards and has now reached values close to replacement level. Focusing on the 1950–70 period, I examine factors that contributed to shaping the Tibetan fertility increase in more detail. This confirms that changes in nuptiality and disease-related infertility both played a role in pushing up fertility rates among Tibetan women in China.     

Behavioral economics and the value of a statistical life

Journal of Risk and Uncertainty - There are many possible connections between the value of statistical life (VSL) and behavioral economics. A list of topics includes endowment effects, risk...     

Challenging the Darwin Skeptics: Examining Proevolutionist Media Persuasion

The socio-religious skirmishes known as the Evolution Wars may have major implications for global religion, education, and popular perceptions of science. These so-called wars have witnessed the rise of several media-vocal proevolutionist organizations, which have sought to challenge Darwin-skeptic ideas and enterprises. Such proevolutionist organizations include the atheistic Richard Dawkins Foundation for Reason and Science, the religiously neutral National Center for Science Education, and the BioLogos Foundation Christian advocacy group. This study executes a content analysis of media produced by these leading proevolutionist organizations from 2009 to 2011 and gauges the occurrence rates of several persuasive cues associated with the Elaboration Likelihood Model. This exercise reveals ways in which certain persuasive elements are articulated throughout proevolutionist media, and juxtaposes these observations with data previously gathered concerning Darwin-skeptic materials.