Markov Chain Monte Carlo in small worlds

As the number of applications for Markov Chain Monte Carlo (MCMC) grows, the power of these methods as well as their shortcomings become more apparent. While MCMC yields an almost automatic way to sample a space according to some distribution, its implementations often fall short of this task as they may lead to chains which converge too slowly or get trapped within one mode of a multi-modal space. Moreover, it may be difficult to determine if a chain is only sampling a certain area of the space or if it has indeed reached stationarity. In this paper, we show how a simple modification of the proposal mechanism results in faster convergence of the chain and helps to circumvent the problems described above. This mechanism, which is based on an idea from the field of “small-world” networks, amounts to adding occasional “wild” proposals to any local proposal scheme. We demonstrate through both theory and extensive simulations, that these new proposal distributions can greatly outperform the traditional local proposals when it comes to exploring complex heterogenous spaces and multi-modal distributions. Our method can easily be applied to most, if not all, problems involving MCMC and unlike many other remedies which improve the performance of MCMC it preserves the simplicity of the underlying algorithm.     

Adaptive sampling without replacement of clusters

In a common form of adaptive cluster sampling, an initial sample of units is selected by random sampling without replacement and, whenever the observed value of the unit is sufficiently high, its neighboring units are added to the sample, with the process of adding neighbors repeated if any of the added units are also high valued. In this way, an initial selection of a high-valued unit results in the addition of the entire network of surrounding high-valued units and some low-valued “edge” units where sampling stops. Repeat selections can occur when more than one initially selected unit is in the same network or when an edge unit is shared by more than one added network. Adaptive sampling without replacement of networks avoids some of this repeat selection by sequentially selecting initial sample units only from the part of the population not already in any selected network. The design proposed in this paper carries this step further by selecting initial units only from the population, exclusive of any previously selected networks or edge units.     

Children, death and ageism

The position of children in the social order can be seen as an ambiguous one. On the one hand, they are highly valued, at a rhetorical level at least, as the basis of the future generation. On the other hand, however, they can be seen to be marginalized and disenfranchised, protected from more aspects of social life than is strictly necessary. This paper explores some of the important issues relating to children's disenfranchised social position, with specific reference to issues relating to death, dying and bereavement.     

The Costs of Future Polio Risk Management Policies

Decisionmakers need information about the anticipated future costs of maintaining polio eradication as a function of the policy options under consideration. Given the large portfolio of options, we reviewed and synthesized the existing cost data relevant to current policies to provide context for future policies. We model the expected future costs of different strategies for continued vaccination, surveillance, and other costs that require significant potential resource commitments. We estimate the costs of different potential policy portfolios for low-, middle-, and high-income countries to demonstrate the variability in these costs. We estimate that a global transition from routine immunization with oral poliovirus vaccine (OPV) to inactivated poliovirus vaccine (IPV) would increase the costs of managing polio globally, although routine IPV use remains less costly than routine OPV use with supplemental immunization activities. The costs of surveillance and a stockpile, while small compared to routine vaccination costs, represent important expenditures to ensure adequate response to potential outbreaks. The uncertainty and sensitivity analyses highlight important uncertainty in the aggregated costs and demonstrates that the discount rate and uncertainty in price and administration cost of IPV drives the expected incremental cost of routine IPV vs. OPV immunization.     

Development and Consideration of Global Policies for Managing the Future Risks of Poliovirus Outbreaks: Insights and Lessons Learned Through Modeling

The success of the Global Polio Eradication Initiative promises to bring large benefits, including sustained improvements in quality of life (i.e., cases of paralytic disease and deaths avoided) and costs saved from cessation of vaccination. Obtaining and maintaining these benefits requires that policymakers manage the transition from the current massive use of oral poliovirus vaccine (OPV) to a world without OPV and free of the risks of potential future reintroductions of live polioviruses. This article describes the analytical journey that began in 2001 with a retrospective case study on polio risk management and led to development of dynamic integrated risk, economic, and decision analysis tools to inform global policies for managing the risks of polio. This analytical journey has provided several key insights and lessons learned that will be useful to future analysts involved in similar complex decision-making processes.     

Machismo: Manifestations of a cultural value in the Latin American casino

Casino operations were observed in twelve Latin American countries. Owners, managers, employees, players, and government regulators in each of these countries were interviewed. The concept of machismo is described in its historical and cultural context. It is then used to illuminate casino operations and the mode of play in existence in these casinos.Research activities for this paper were supported by grants from the First Interstate Bank of Las Vegas, Nevada, through the Center for Business Leadership at the University of Nevada, Las Vegas, and from Tottenham and Associates, Gaming Consultants, London, England.     

Validating Benefit and Cost Estimates: The Case of Airbag Regulation

Preregulation estimates of benefits and costs are rarely validated after regulations are implemented. This article performs such a validation for the mandatory automobile airbag requirement. We found that the original 1984 model used to estimate benefits became invalid when 1997 values were input into that 1984 model. However, using a published 1997 cost-effectiveness model, we demonstrate, by replacing the model inputs with the values from 1984, that the 1997 cost-effectiveness ratios, based on real-world crash data and tear-down cost data, are less attractive than what would have been originally anticipated. The three most important errors in the 1984 input values are identified: the overestimation of airbag effectiveness, the overestimation of baseline fatality/injury rates, and the underestimation of manual safety belt use. This case study, which suggests that airbags are a reasonable investment in safety, shows that the regulatory analysis tools do not always produce findings that are biased against health, safety, and environmental regulation. Future validation studies of health, safety, and environmental regulation should focus on validation of benefit and risk estimates, areas where we found significant error, as well as on cost estimates.     

Likelihood Ratio Tests for Fixed Model Terms using Residual Maximum Likelihood

Likelihood ratio tests for fixed model terms are proposed for the analysis of linear mixed models when using residual maximum likelihood estimation. Bartlett-type adjustments, using an approximate decomposition of the data, are developed for the test statistics. A simulation study is used to compare properties of the test statistics proposed, with or without adjustment, with a Wald test. A proposed test statistic constructed by dropping fixed terms from the full fixed model is shown to give a better approximation to the asymptotic χ²-distribution than the Wald test for small data sets. Bartlett adjustment is shown to improve the χ²-approximation for the proposed tests substantially.     

Aetiology of anorexia nervosa in Hong Kong: a social work qualitative inquiry

This paper reports the results of a qualitative inquiry into the aetiology of anorexia nervosa in Hong Kong. These results are part of our main research project about family treatment and families with a daughter suffering from anorexia nervosa. This present qualitative study adopts a multiple case study approach (n = 8) and employs a multiple‐source data collection method, involving pre‐ and post‐family treatment interviews, clinical observations of the participants during the treatment sessions and home visits, as well as detailed field notes. The findings indicate that there are biological, psychological, socio‐cultural and familial interpretations of the aetiology of anorexia nervosa. This study challenges the dominant discourse on anorexia nervosa in Hong Kong, which states that the root of the disease lies in a desire for slenderness and beauty, and suggests instead that the aetiology of anorexia nervosa can be discovered only if approached from the systems level. Implications for social work practice are highlighted.