The investigator studies and develops computer-intensive methods for applied statistics, with applicatiion in biology. His current proposal has three projects a) complementary clustering, a kind of orthogonal decomposition analogous to principal compoennts, b) the fused lasso for spatial smoothing and hot-spot detection, c) pre-validation- a method for inference for the p>N setting and d) The new edition of the text "The Elements of Statistical Learning"
There have been significant developments in the areas of applied regression and classification over the past 10-15 years. Much of the impetus originally came from outside of the field of statistics, from areas such as computer science, machine learning and neural networks. As a result, we now have at our disposal a very powerful collection of techniques for adaptive regression and classification. These are now being applied to medical diagnosis, bioinformatics and genetic modeling, chemical process control, shape, handwriting, speech and face recognition, financial modeling, and a wide range of other important practical problems. In this work the investigator plans to develop and study new tools for the important practical problems.
