Blocking in two-level non-regular fractional factorial designs

In this paper we consider screening experiments where a two-level fractional factorial design is to be used to identify significant factors in an experimental process and where the runs in the experiment are to occur in blocks of equal size. A simple method based on the foldover technique is given for constructing resolution IV orthogonal and non-orthogonal blocked designs and examples are given to illustrate the process.     

The social inclusion policy agenda in Australia: a case of old wine,new bottles?

The election of an Australian Labor Government in Australia in 2007 saw ‘social inclusion’ emerge as the official and overarching social policy agenda. Being ‘included’ was subsequently defined by the ALP Government as being able to ‘have the resources, opportunities and capabilities needed to learn, work, engage and have a voice‘. Various researchers in Australia demonstrated an interest in social inclusion, as it enabled them to construct a multi‐dimensional framework for measuring disadvantage. This research program resulted in various forms of statistical modelling based on some agreement about what it means to be included in society. The multi‐dimensional approach taken by academic researchers, however, did not necessarily translate to a new model of social policy development or implementation. We argue that, similar to the experience of the UK, Australia's social inclusion policy agenda was for the most part narrowly and individually defined by politicians and policy makers, particularly in terms of equating being employed with being included. We conclude with discussion about the need to strengthen the social inclusion framework by adopting an understanding of social inequality and social justice that is more relational and less categorical.     

Missing data: Discussion points from the PSI missing data expert group

The Points to Consider Document on Missing Data was adopted by the Committee of Health and Medicinal Products (CHMP) in December 2001. In September 2007 the CHMP issued a recommendation to review the document, with particular emphasis on summarizing and critically appraising the pattern of drop‐outs, explaining the role and limitations of the ‘last observation carried forward’ method and describing the CHMP's cautionary stance on the use of mixed models. In preparation for the release of the updated guidance document, statisticians in the Pharmaceutical Industry held a one‐day expert group meeting in September 2008. Topics that were debated included minimizing the extent of missing data and understanding the missing data mechanism, defining the principles for handling missing data and understanding the assumptions underlying different analysis methods. A clear message from the meeting was that at present, biostatisticians tend only to react to missing data. Limited pro‐active planning is undertaken when designing clinical trials. Missing data mechanisms for a trial need to be considered during the planning phase and the impact on the objectives assessed. Another area for improvement is in the understanding of the pattern of missing data observed during a trial and thus the missing data mechanism via the plotting of data; for example, use of Kaplan–Meier curves looking at time to withdrawal. Copyright © 2009 John Wiley & Sons, Ltd.     

High-Dimensional Sparse Factor Modeling: Applications in Gene Expression Genomics

We describe studies in molecular profiling and biological pathway analysis that use sparse latent factor and regression models for microarray gene expression data. We discuss breast cancer applications and key aspects of the modeling and computational methodology. Our case studies aim to investigate and characterize heterogeneity of structure related to specific oncogenic pathways, as well as links between aggregate patterns in gene expression profiles and clinical biomarkers. Based on the metaphor of statistically derived "factors" as representing biological "subpathway" structure, we explore the decomposition of fitted sparse factor models into pathway subcomponents and investigate how these components overlay multiple aspects of known biological activity. Our methodology is based on sparsity modeling of multivariate regression, ANOVA, and latent factor models, as well as a class of models that combines all components. Hierarchical sparsity priors address questions of dimension reduction and multiple comparisons, as well as scalability of the methodology. The models include practically relevant non-Gaussian/nonparametric components for latent structure, underlying often quite complex non-Gaussianity in multivariate expression patterns. Model search and fitting are addressed through stochastic simulation and evolutionary stochastic search methods that are exemplified in the oncogenic pathway studies. Supplementary supporting material provides more details of the applications, as well as examples of the use of freely available software tools for implementing the methodology.     

Abuse of Adolescents: Implications of North American Research for the UK

In the last edition of this journal we reviewed the extensive North Amercican (mainly US) literature on the abuse of adolescents. Here we look at the implications of the research for the current UK context.     

Survey design for detecting patches

In this paper, we are concerned with designing surveys for detecting patches of some exploitable resource, such as a shellfish stock. We take as our objective that, if there is one or more circular patches above some specified size, then the probability of detecting at least one patch should be high. We show how the required sampling intensity can be modified in the light of information about the likely number and size of patches. The results are applied to two surveys for detecting exploitable patches of cockles ( Cerastoderma edule ).     

ESTIMATING EQUATIONS AND NON-LINEAR FUNCTIONAL RELATIONSHIPS

A nonlinear functional relationship is defined as an R-dimensional manifold in P-dimensional space. The formulation of the model may be explicitly in terms of R-dimensional vectors of incidental parameters or implicitly by a (P-R)-dimensional vector function of constraints. The objective is to estimate and make inference about a K-vector of parameters θ which defines the manifold. Each observed P-vector has its expectation lying on the manifold, and the error vector has a variance matrix defined in terms of a further vector of parameters The theory of estimating equations in the presence of incidental parameters is extended and applied to the explicit formulation, to give equations suitable for estimating θ given knowledge of only the first two moments. The method has a geometrical interpretation. Estimating equations for are chosen to be those which would be optimal if the normality assumption were true. First order corrections to the biases of these estimates are included. An example where the manifold is a circle centred on the origin is used to illustrate the theory. Further examples incorporate more general features, including the estimation of two variance parameters and estimation in higher dimensions.