In classical statistical data analysis, we are used to dealing with just a handful of carefully chosen variables. However, in many contemporary scientific problems and in commercial settings too, technological advances have meant that we can collect a large number of predictors with which we would hope to explain a given phenomenon of interest. High-throughput data in genomics, click-through rate prediction for internet advertising and large-scale databases in healthcare analytics are just a few of the myriad settings in which high-dimensional data are encountered routinely. The field of high-dimensional statistics is a response to the challenges posed by these sorts of data, and is one of the most active areas of statistics on the international stage.
Our aims in this Interest Group are to keep abreast of the rapid developments in the field by discussing important recent papers related to methodological aspects of high-dimensional statistics (broadly interpreted), and to identify some key application-driven problems for which methodology is currently unavailable.
Details of upcoming workshops and meetings will be shared soon.