Free energy methods for Bayesian inference: efficient exploration of univariate Gaussian mixture posteriors

Because of their multimodality, mixture posterior distributions are difficult to sample with standard Markov chain Monte Carlo (MCMC) methods. We propose a strategy to enhance the sampling of MCMC in this context, using a biasing procedure which originates from computational Statistical Physics. The principle is first to choose a “reaction coordinate”, that is, a “direction” in which the target distribution is multimodal. In a second step, the marginal log-density of the reaction coordinate with respect to the posterior distribution is estimated; minus this quantity is called “free energy” in the computational Statistical Physics literature. To this end, we use adaptive biasing Markov chain algorithms which adapt their targeted invariant distribution on the fly, in order to overcome sampling barriers along the chosen reaction coordinate. Finally, we perform an importance sampling step in order to remove the bias and recover the true posterior. The efficiency factor of the importance sampling step can easily be estimated a priori once the bias is known, and appears to be rather large for the test cases we considered.     

Identifying Male College Students' Perceived Health Needs,Barriers to Seeking Help,and Recommendations to Help Men Adopt Healthier Lifestyles

Abstract

Seven focus groups at a university campus were formed to identify college men's health concerns, barriers to seeking help, and recommendations to help college men adopt healthier lifestyles. Content analysis was used to identify and organize primary patterns in the focus-group data. Results of the study revealed that the college men were aware that they had important health needs but took little action to address them. The participants identified both physical and emotional health concerns. Alcohol and substance abuse were rated as the most important issues for men. The greatest barrier to seeking services was the men's socialization to be independent and conceal vulnerability. The most frequently mentioned suggestions for helping men adopt healthier lifestyles were offering health classes, providing health information call-in service, and developing a men's center. Implications of the results are discussed.     

North Americans' Attitudes Toward Illegal Drugs

Public opinion polls conducted from 1969 to 2003 found a variety of opinions on the use, possession, and legalization of marijuana. Public opinion about legalizing marijuana use is increasing, but it is also increasing toward harsher penalties for those who possess a small amount of marijuana. Public opinion regarding the legalization of illegal drugs appears to be influenced by the times. The majority of Americans increasingly believe that the country has made some progress in dealing with the problems of illegal drugs, but Americans are also reporting more difficulty within their families related to marijuana. This article examines trends based in public opinion polls on the use of marijuana in the United States and Canada and discusses the social and political history of marijuana and the biophysical effects, medical uses, definitions, and previous research related to marijuana.     

Impulsivity and pathological gambling across depressive versus bipolar disorders

Impulsivity has been implicated in the development of pathological gambling (PG); sensation seeking and urgency in particular have predicted gambling pathology in undergraduate and psychiatric samples. In light of the relevance of both depressed and elevated mood to impulsivity and gambling, the components of impulsivity associated with PG across mood disorders warrants investigation. The aim of the current investigation was to examine the association between impulsivity and gambling pathology severity across depressive versus bipolar disorders. A total of 275 participants with lifetime depressive or bipolar disorder completed measures of impulsivity and gambling. Urgency was consistently associated with gambling pathology indicators; lack of perseverance was specifically associated with gambling pathology within participants with depressive disorders. Reckless action during negative mood is associated with gambling pathology across mood disorders, whereas difficulty remaining focused is associated with PG solely within depressive disorders. Impulsivity and affective comorbidity may inform current understanding of PG.     

Thinking about a new social work curriculum

This article reports the findings of a research study into the responses of practice teachers to their recently enhanced responsibilities in providing learning opportunities to, and assessing the competence of, students to practise social work law. The article examines their familiarity with social work law, including their training, knowledge base, resources and employer support. A critical analysis is offered of the extent to which CCETSW's requirements on teaching social work law in placements are being realised.     

‘Closing the Gap’ in Indigenous life expectancies: what if we succeed?

In 2007–2008 the Council of Australian Governments agreed to six ambitious targets for addressing longstanding disparities between Indigenous and other Australians in health, education and employment outcomes. The ‘National Integrated Strategy for Closing the Gap on Indigenous Disadvantage’ (colloquially ‘Closing the Gap’) includes the goal of eliminating life expectancy gaps within a generation. This policy says nothing about what changes in the demographic structure of the population might be expected should there be even partial success towards this ambitious target. Information is required to analyse age and sex shifts for their effects on service demand and provision. In this study we apply cohort component projections modelling to the Northern Territory, the Australian jurisdiction with the largest Indigenous component in its population and the largest life expectancy gaps, to assess the demographic effects of closing the gap within a generation. Three scenarios are modelled: (1) No changes to Indigenous life expectancies from those estimated in 2010; (2) Complete success in closing the gap within a generation; and (3) A continuation of current forecasts about how Indigenous life expectancies will change into the future. Although closing the gap would only produce a small increase in the size of the next generation’s Indigenous population, over and above that projected with lower life expectancy scenarios, it would result in substantial changes in age compositions, with associated shifts in various demographic indices. These are pertinent to the planning and funding of core services into the future.     

Clarifying Types of Uncertainty: When Are Models Accurate,and Uncertainties Small?

Professor Aven has recently noted the importance of clarifying the meaning of terms such as "scientific uncertainty" for use in risk management and policy decisions, such as when to trigger application of the precautionary principle. This comment examines some fundamental conceptual challenges for efforts to define "accurate" models and "small" input uncertainties by showing that increasing uncertainty in model inputs may reduce uncertainty in model outputs; that even correct models with "small" input uncertainties need not yield accurate or useful predictions for quantities of interest in risk management (such as the duration of an epidemic); and that accurate predictive models need not be accurate causal models.     

A new variant of dual-record population estimation with an application in remote indigenous communities

Dual-record system methods are commonly used as a basis for population estimation. A basic assumption is that the units sampled are drawn only from the population to be estimated. This assumption cannot be met for remote Indigenous communities in Australia. A new variant of dual-record population estimation is presented, which relies on the availability of specific additional information to relax the assumption of perfect frame specification. This variant is applied to two remote Indigenous communities in the Northern Territory of Australia, using locally available data sources. Further theoretical exploration of this method is presented along with possible applications in estimating area-enumerated populations and census coverage.