Sparse estimates in GLM with environmental application

Warning

This publication doesn't include Institute of Computer Science. It includes Faculty of Economics and Administration. Official publication website can be found on muni.cz.

Authors	VESELÝ Vítězslav HÜBNEROVÁ Zuzana MICHÁLEK Jaroslav
Year of publication	2009
Type	Conference abstract
MU Faculty or unit	Faculty of Economics and Administration
Citation
Description	We give some minimal theoretical background of the sparse estimation technique based on the four-step Basis Pursuit Algorithm (BPA4) and sketch briefly its steps. The new concept is illustrated by two application examples: (1) simulation study proving better numerical stability in contrast with standard estimation methods, (2) strongly rank-deficient environmental GLM of air pollution by suspended particulate matter.
Related projects:	Dynamic Geovisualisation in Crises Management