Automatic Synthesis of Statistical Data Analysis Programs Position Statement m

Statistical data analysis is a core activity in experimen-tal sciences. It encompasses a wide variety of tasks, rang-ing from, e.g., a simple linear regression to fitting com-plex dynamical models. Developing statistical data analysis programs, however, is an arduous and error-prone process which requires profound expertise in different areas: statis-tics, numerics, software engineering, and of course the sci-entific application domain. We believe that statistical data analysis is a very promis-ing domain for the application program synthesis, despite--or even because--these difficulties. Statistics provides a uni-fying and concise domain-specific notation. Graphical mod-els (Buntine 1994) provide a structuring mechanism which can be exploited during the synthesis process, e.g., to de-compose a problem into independent subproblems. Statisti