Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 2: Poster
SOE 2.3: Poster
Wednesday, September 29, 2021, 17:00–18:00, P
The Principle of Largest Squares — •Martin Erik Horn — IU Internatinal University of Applied Sciences, Campus Berlin, and ISM International School of Management, Campus Berlin
Regression analysis is an important statistical tool to understand interdependencies between different variables in empirical sciences. And it is astonishing that this tool obviously is used only in Euclidean spaces - as if variables always have to act in an Euclidean way in the scientific world of socio economics and of other domains.
To critically question this Euclidean dominance, orthogonal regression will be transferred into spacetime. The didactical consequences are interesting: We are then discussing relativistic ideas with students who do not study physics, but economics, computer science or other subjects relying on empirical analysis. And the conceptual consequences are surprising: We will no longer apply the principle of least squares but have to switch to the principle of largest squares.