University of Cambridge > Talks.cam > Statistics > Invariant Coordinate Selection revisited: Fisher Symmetry and Symmetric Component Analysis

Invariant Coordinate Selection revisited: Fisher Symmetry and Symmetric Component Analysis

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact .

Tyler et al. (2009) introduced invariant coordinate selection, or ICS , as a general method for exploring affine invariant features of multivariate data by comparing different estimates of multivariate scatter. Together with Critchley et al. (2006), they report examples of the method performing well for a wide range of problems, extending beyond the limits of existing theoretical support. Motivated by this, we provide complementary ICS theory based on the relevant symmetry group. A Fisher symmetry condition is introduced for which elliptical symmetry is not required, yet under which a subset of the invariant coordinates is shown to correspond to Fisher’s linear discriminant subspace, class identifications of data points remaining unknown. Again, a Symmetric Component Analysis model is introduced in which independence is not required, yet under which the invariant coordinates are seen to correspond to the symmetric components. Illustrative examples are given. Further developments are briefly indicated.

This talk is part of the Statistics series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity