Using Lisp Macro Facilities for Transferable Statistical Tests

Описание: ELS 2016, 9th European Lisp Symposium, 9-10 May 2016,
Department of Computer Science, AGH University of Science and Technology, Kraków, Poland.

Kay Hamacher, Dept. of Computer Science, Dept. of Physics, Dept. of Biology Darmstadt, Germany.

Model-free statistical tests are purely data-driven approaches to assess correlations and other interdependencies between observable quantities. The few, distinct patterns how to perform these tests on the myriad of potentially different interdependence measures prompted us to use (Common) Lisp’s macro capabilities for the development of a general, domain-specific language (DSL) of expectation values under so-called resampling techniques. Herein, we give an intro- duction into this statistical approach to big data, describe our solution, and report on application as well as on further research opportunities in statistical DSLs. We illustrate the results based on a toy example.