Boost Your Spatiotemporal Data Analysis with Physical Knowledge

• Activity Number: 95
• Type: Roundtables
• Date/Time: Monday, August 5, 2013 : 7:00 AM to 8:15 AM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Alexander Kolovos*+
• Companies: SpaceTimeWorks, LLC
• Keywords: Spatiotemporal ; spatial ; prediction ; modeling ; physical knowledge ; interdisciplinary

Abstract

Spatial and space-time statistics commonly employ a strictly data-based approach for the purpose of inference and simulation. Related methodologies explore interdependencies among observed data to provide explanatory mechanisms for the attributes of interest. Yet, in many of the problems encountered in physical and engineering sciences, your research might have a distinct advantage by means of physical principles, models, and laws that possibly apply and could improve your data-based analysis. The present roundtable discussion starts from recent publications in spatiotemporal analysis topics, where such additional knowledge bases have been employed to strengthen the analytical foundation and improve the results. Examples include spatiotemporal covariances and complementary data derived from relevant physical models. Join us to get ideas about applying related principles in your research and to discuss your own approaches when you cope with similar analyses. Our roundtable celebrates the International Year of Statistics by bringing together the statistics behind spatiotemporal data research with the physics governing the natural processes that produce those data.