Geodesic least squares regression for scaling studies in magnetic confinement fusion
- Author
- Geert Verdoolaege (UGent)
- Organization
- Abstract
- In regression analyses for deriving scaling laws that occur in various scientific disciplines, usually standard regression methods have been applied, of which ordinary least squares (OLS) is the most popular. However, concerns have been raised with respect to several assumptions underlying OLS in its application to scaling laws. We here discuss a new regression method that is robust in the presence of significant uncertainty on both the data and the regression model. The method, which we call geodesic least squares regression (GLS), is based on minimization of the Rao geodesic distance on a probabilistic manifold. We demonstrate the superiority of the method using synthetic data and we present an application to the scaling law for the power threshold for the transition to the high confinement regime in magnetic confinement fusion devices.
- Keywords
- DISTANCE, MODELS, scaling laws, regression, nuclear fusion, information geometry, geodesic distance
Downloads
-
(...).pdf
- full text
- |
- UGent only
- |
- |
- 464.00 KB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-6871849
- MLA
- Verdoolaege, Geert. “Geodesic Least Squares Regression for Scaling Studies in Magnetic Confinement Fusion.” AIP Conference Proceedings, edited by A MohammadDjafari and F Barbaresco, vol. 1641, 2015, pp. 564–71, doi:10.1063/1.4906023.
- APA
- Verdoolaege, G. (2015). Geodesic least squares regression for scaling studies in magnetic confinement fusion. In A. MohammadDjafari & F. Barbaresco (Eds.), AIP Conference Proceedings (Vol. 1641, pp. 564–571). https://doi.org/10.1063/1.4906023
- Chicago author-date
- Verdoolaege, Geert. 2015. “Geodesic Least Squares Regression for Scaling Studies in Magnetic Confinement Fusion.” In AIP Conference Proceedings, edited by A MohammadDjafari and F Barbaresco, 1641:564–71. https://doi.org/10.1063/1.4906023.
- Chicago author-date (all authors)
- Verdoolaege, Geert. 2015. “Geodesic Least Squares Regression for Scaling Studies in Magnetic Confinement Fusion.” In AIP Conference Proceedings, ed by. A MohammadDjafari and F Barbaresco, 1641:564–571. doi:10.1063/1.4906023.
- Vancouver
- 1.Verdoolaege G. Geodesic least squares regression for scaling studies in magnetic confinement fusion. In: MohammadDjafari A, Barbaresco F, editors. AIP Conference Proceedings. 2015. p. 564–71.
- IEEE
- [1]G. Verdoolaege, “Geodesic least squares regression for scaling studies in magnetic confinement fusion,” in AIP Conference Proceedings, Amboise, France, 2015, vol. 1641, pp. 564–571.
@inproceedings{6871849, abstract = {{In regression analyses for deriving scaling laws that occur in various scientific disciplines, usually standard regression methods have been applied, of which ordinary least squares (OLS) is the most popular. However, concerns have been raised with respect to several assumptions underlying OLS in its application to scaling laws. We here discuss a new regression method that is robust in the presence of significant uncertainty on both the data and the regression model. The method, which we call geodesic least squares regression (GLS), is based on minimization of the Rao geodesic distance on a probabilistic manifold. We demonstrate the superiority of the method using synthetic data and we present an application to the scaling law for the power threshold for the transition to the high confinement regime in magnetic confinement fusion devices.}}, author = {{Verdoolaege, Geert}}, booktitle = {{AIP Conference Proceedings}}, editor = {{MohammadDjafari, A and Barbaresco, F}}, isbn = {{9780735412804}}, issn = {{0094-243X}}, keywords = {{DISTANCE,MODELS,scaling laws,regression,nuclear fusion,information geometry,geodesic distance}}, language = {{eng}}, location = {{Amboise, France}}, pages = {{564--571}}, title = {{Geodesic least squares regression for scaling studies in magnetic confinement fusion}}, url = {{http://doi.org/10.1063/1.4906023}}, volume = {{1641}}, year = {{2015}}, }
- Altmetric
- View in Altmetric
- Web of Science
- Times cited: