A Bayesian framework to update scaling factors for radioactive waste characterization.

Nuclear power plants and research facilities commonly employ the so-called scaling factor (SF) method to quantify the activity of difficult-to-measure (DTM) radionuclides within their radioactive waste packages. The method relies on the establishment of a relationship between an easy-to-measure (ETM) radionuclide, called key nuclide (KN), and difficult-to-measure radionuclides, after the collection of a representative sample from the waste population. The distribution of the scaling factors, as well as the parameters defining the distribution, can change over time. Therefore, the accuracy of the calculated activity of the DTM radionuclides depends on the capacity of the scaling factor method to follow the time evolution of the waste population. In practice, waste producers collect periodically new samples from the waste population and check the variation and the validity of the scaling factors. In this article, we present a simple Bayesian framework to update scaling factors when a new data set becomes available. The method is tested and validated for radioactive waste produced at CERN (European Organization for Nuclear Research) and can be easily implemented for waste of different origin.

A robust approach for skewed and heavy-tailed outcomes in the analysis of health care expenditures.

In this paper robust statistical procedures are presented for the analysis of skewed and heavy-tailed outcomes as they typically occur in health care data. The new estimators and test statistics are extensions of classical maximum likelihood techniques for generalized linear models. In contrast to their classical counterparts, the new robust techniques show lower variability and excellent efficiency properties in the presence of small deviations from the assumed model, i.e. when the underlying distribution of the data lies in a neighborhood of the model. A simulation study, an analysis on real data, and a sensitivity analysis confirm the good theoretical statistical properties of the new techniques.

Variable selection for marginal longitudinal generalized linear models.

Variable selection is an essential part of any statistical analysis and yet has been somewhat neglected in the context of longitudinal data analysis. In this article, we propose a generalized version of Mallows's C(p) (GC(p)) suitable for use with both parametric and nonparametric models. GC(p) provides an estimate of a measure of model's adequacy for prediction. We examine its performance with popular marginal longitudinal models (fitted using GEE) and contrast results with what is typically done in practice: variable selection based on Wald-type or score-type tests. An application to real data further demonstrates the merits of our approach while at the same time emphasizing some important robust features inherent to GC(p).