- Statist. Sci.
- Volume 28, Number 1 (2013), 1-39.
Elliptical Insights: Understanding Statistical Methods through Elliptical Geometry
Visual insights into a wide variety of statistical methods, for both didactic and data analytic purposes, can often be achieved through geometric diagrams and geometrically based statistical graphs. This paper extols and illustrates the virtues of the ellipse and her higher-dimensional cousins for both these purposes in a variety of contexts, including linear models, multivariate linear models and mixed-effect models. We emphasize the strong relationships among statistical methods, matrix-algebraic solutions and geometry that can often be easily understood in terms of ellipses.
Statist. Sci., Volume 28, Number 1 (2013), 1-39.
First available in Project Euclid: 29 January 2013
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Added-variable plots Bayesian estimation concentration ellipse data ellipse discriminant analysis Francis Galton hypothesis-error plots kissing ellipsoids measurement error mixed-effect models multivariate meta-analysis regression paradoxes ridge regression statistical geometry
Friendly, Michael; Monette, Georges; Fox, John. Elliptical Insights: Understanding Statistical Methods through Elliptical Geometry. Statist. Sci. 28 (2013), no. 1, 1--39. doi:10.1214/12-STS402. https://projecteuclid.org/euclid.ss/1359468407
- Supplementary material: Supplementary materials for Elliptical insights: Understanding statistical methods through elliptical geometry. The supplementary materials include SAS and R scripts to generate all of the figures for this article. Several 3D movies are also included to show phenomena better than can be rendered in static print images. A new R package, gellipsoid, provides computational support for the theory described in Appendix A.1. These are also available at http://datavis.ca/papers/ellipses and described in http://datavis.ca/papers/ellipses/supp.pdf [Friendly, Monette, Fox (2012)].