Reml and best linear unbiased prediction in state space models

This article takes a hierarchical model approach to the estimation of state space models with diffuse initial conditions. An initial state is said to be diffuse when it cannot be assigned a proper prior distribution. In state space models this occurs either when fixed effects are present or when modelling nonstationarity in the state transition equation. Whereas much of the literature views diffuse states as an initialization problem, we follow the approach of Sallas and Harville (1981,1988) and incorporate diffuse initial conditions via noninformative prior distributions into hierarchical linear models. We apply existing results to derive the restricted loglike-lihood and appropriate modifications to the standard Kalman filter and smoother. Our approach results in a better understanding of De Jong's (1991) contributions. This article also shows how to adjust the standard Kalman filter, the fixed inter- val smoother and the state space model forecasting recursions, together with their mean square errors, for he presence of diffuse components. Using a hierarchical model approach it is shown that the estimates obtained are Best Linear Unbiased Predictors (BLUP).     

Disability and Dependency: origins and futures of 'special needs' housing for disabled people

In this paper we analyse the provision of housing for disabled people in both its historical and ideological contexts. While the recent extension of part M of the Building Regulations to dwellings by the government represents a significant advance towards the inclusion of disabled people, we argue that the shift in funding of public housing from a 'bricks and mortar' subsidy to Housing Benefit potentially creates greater dependency. This amounts to a policy contradiction in which the government is punishing disabled people for being dependent whilst simultaneously being the cause of their dependency. If this issue is not addressed disabled people will continue to remain excluded, albeit within accessible dwellings.     

The robustness of the quasilikelihood estimator

The quasilikelihood estimator is widely used in data analysis where a likelihood is not available. We illustrate that with a given variance function it is not only conservative, in minimizing a maximum risk, but also robust against a possible misspecification of either the likelihood or cumulants of the model. In examples it is compared with estimators based on maximum likelihood and quadratic estimating functions.     

Gender Differences in Personal Income and Financial Risk Tolerance: How Much of a Connection?

Prior research indicates that relative to men, women generally have lower incomes and typically are more risk averse. In a sample of the clientele of financial planners, men had higher personal incomes and exhibited greater financial risk tolerance, which is consistent with previous studies. The authors' objective was to determine how much of the gender difference in income could be explained by risk tolerance. Results indicate that risk tolerance can explain some of the variation in earnings between women and men but that it is not a primary reason for the wage gap.     

Quantitative Risk Assessment of the New York State Operated West Valley Radioactive Waste Disposal Area

This article is based on a quantitative risk assessment (QRA) that was performed on a radioactive waste disposal area within the Western New York Nuclear Service Center in western New York State. The QRA results were instrumental in the decision by the New York State Energy Research and Development Authority to support a strategy of in‐place management of the disposal area for another decade. The QRA methodology adopted for this first of a kind application was a scenario‐based approach in the framework of the triplet definition of risk (scenarios, likelihoods, consequences). The measure of risk is the frequency of occurrence of different levels of radiation dose to humans at prescribed locations. The risk from each scenario is determined by (1) the frequency of disruptive events or natural processes that cause a release of radioactive materials from the disposal area; (2) the physical form, quantity, and radionuclide content of the material that is released during each scenario; (3) distribution, dilution, and deposition of the released materials throughout the environment surrounding the disposal area; and (4) public exposure to the distributed material and the accumulated radiation dose from that exposure. The risks of the individual scenarios are assembled into a representation of the risk from the disposal area. In addition to quantifying the total risk to the public, the analysis ranks the importance of each contributing scenario, which facilitates taking corrective actions and implementing effective risk management. Perhaps most importantly, quantification of the uncertainties is an intrinsic part of the risk results. This approach to safety analysis has demonstrated many advantages of applying QRA principles to assessing the risk of facilities involving hazardous materials.     

Institutional Diversity and Performance in African Cotton Sectors

This article analyses the performance of cotton sectors across East, Southern, and West Africa, paying particular attention to the wide diversity of institutional arrangements that they now exhibit. It finds strong support for earlier contentions regarding trade‐offs between competition and coordination, and between the roles of public and private sectors. New insights provide concrete and context‐specific guidance to policy‐makers and stakeholders regarding the key challenges they will face and the risks they will need to manage as they work to improve productivity and ensure an equitable division of benefits within cotton sectors.     

Temporal Relationships Among Depressive Symptoms,Risky Behavior Engagement,Perceived Control,and Gender in a Sample of Adolescents

The goals of the study were to examine whether (a) risky behaviors precede depressive symptoms or, conversely, depressive symptoms predict risky behavior engagement; (b) gender moderates the relationship between risky behavior engagement and depressive symptoms; and (c) perceived control strengthens the association between risky behavior engagement and depressive symptoms. At Time 1, 118 adolescents completed self‐report measures assessing perceived control, risky behavior engagement, and depressive symptoms. Follow‐up assessments occurred every 6 weeks (Times 2–5), and participants completed measures assessing risky behavior engagement and depressive symptoms. Results indicated that neither risky behavior engagement nor depressive symptoms emerged as main effects for the sample as a whole. When examining the relationship between depressive symptoms and risky behavior engagement as a function of gender, boys', but not girls', risky behavior engagement predicted higher levels of depressive symptoms. Irrespective of whether we examined boys or girls, depressive symptoms did not predict risky behavior engagement. With regards to the role of cognitive vulnerability, adolescents with lower levels of perceived control reported higher levels of depressive symptoms following risky behavior engagement. These findings suggest that both gender and cognitive vulnerability factors may potentiate the relationship between risky behavior engagement and subsequent depressive symptoms.