Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
711153 | IFAC-PapersOnLine | 2015 | 6 Pages |
Abstract
The question of when estimated correlations should be considered significant is studied. It is illustrated that the established praxis for detecting significant correlations is misleading and cannot be used for automatic model validation. Instead, a modification is introduced and clearly seen to be a significant improvement to the established praxis.
Related Topics
Physical Sciences and Engineering
Engineering
Computational Mechanics