FocuStat: Focus Driven Statistical Inference with Complex Data (completed)
FocuStat is a five-year project funded by the Research Council of Norway, operating from January 2014 to December 2018. The project group consists of Professor Nils Lid Hjort (project leader), Gudmund Hermansen and Kristoffer Hellton (PostDocs), Sam-Erik Walker and Céline Cunen (PhDs). Other PhD and Master's level students are also associated with the project (see the Who We Are list); in particular, PhD students Vinnie Ko and Emil Aas Stoltenberg are active participants and take part in weekly group meetings. Also, check out our Facebook page, with photos and informal reports from conferences, workshops, research kitchens, some general discussion, etc.
About the project
Statistics is the science of reaching decisions under uncertainty and is in many respects a far-ranging success story, permeating nearly all substantive sciences and areas of society where data are collected. It has used around hundred years to reach its present state of high maturity and uniform usefulness. In broad strokes, the four main areas associated with
- parametrics (models indexed by low-dimensional parameters)
- nonparametrics (models with high- or infinite-dimensional parameters)
- assessment and selection of models
- combination of different sources of information
drive most of modern statistics, and have, in essence, been well sorted out, conceptually and operationally. There are important gaps to be filled and new paradigms and principles to develop, however, when faced with the statistical challenges of the 21st century. New types of data, related both to new types of substantive questions in a changing society and to evolving technologies for monitoring and examining more complicated phenomena than earlier, create a need for new types of statistical modelling for new types of analysis, and potentially also for new concepts of information and inference. Themes and challenges we are working on share the concept of the focus, the operational view that the science and context drive the most important questions which again should influence the optimal combinations of models, their analysis, and the ensuing decisions. Three such challenges are as follows:
- A: "breaking the wall" between areas (1) and (2), partly leaning on recent advances inside area (3).
- B: extending the current scope and catalogue of approaches and methods of relevance to area (4), including the use of confidence distributions.
- C: extending and developing new methodologies for areas (1) and (3) for the by now frequently occurring situations where the number of measurements per individual exceeds the sample size.
The project is funded by the Research Council of Norway.