Invariant Coordinate Selection
نویسندگان
چکیده
A general method for exploring multivariate data by comparing different estimates of multivariate scatter is presented. The method is based upon the eigenvalue-eigenvector decomposition of one scatter matrix relative to another. In particular, it is shown that the eigenvectors can be used to generate an affine invariant coordinate system for the multivariate data. Consequently, we view this method as a method for invariant coordinate selection (ICS). By plotting the data with respect to this new invariant coordinate system, various data structures can be revealed. For example, under certain independent components models, it is shown that the invariant coordinates correspond to the independent components. Another example pertains to mixtures of elliptical distributions. In this case, it is shown that a subset of the invariant coordinates corresponds to Fisher’s linear discriminant subspace, even though the class identifications of the data points are unknown. Some illustrative examples are given.
منابع مشابه
Coordinate finite type invariant surfaces in Sol spaces
In the present paper, we study surfaces invariant under the 1-parameter subgroup in Sol space $rm Sol_3$. Also, we characterize the surfaces in $rm Sol_3$ whose coordinate functions of an immersion of the surface are eigenfunctions of the Laplacian $Delta$ of the surface.
متن کاملInvariant coordinate selection for multivariate data analysis - the package ICS
Invariant coordinate selection (ICS ) has recently been introduced by Tyler et al. (2008) as a method for exploring multivariate data. It includes as shown in Oja et al. (2006) as a special case a method for recovering the unmixing matrix in independent components analysis (ICA). It also serves as a basis for classes of multivariate nonparametric tests. The aim of this paper is to briefly expla...
متن کاملPenalized Bregman Divergence Estimation via Coordinate Descent
Variable selection via penalized estimation is appealing for dimension reduction. For penalized linear regression, Efron, et al. (2004) introduced the LARS algorithm. Recently, the coordinate descent (CD) algorithm was developed by Friedman, et al. (2007) for penalized linear regression and penalized logistic regression and was shown to gain computational superiority. This paper explores...
متن کاملDynamics of Differential Equations on Invariant Manifolds
The simplification resulting from reduction of dimension involved in the study of invariant manifolds of differential equations is often difficult to achieve in practice. Appropriate coordinate systems are difficult to find or are essentially local in nature thus complicating analysis of global dynamics. This paper develops an approach which avoids the selection of coordinate systems on the man...
متن کاملOutput Consensus Control of Nonlinear Non-minimum Phase Multi-agent Systems Using Output Redefinition Method
This paper concerns the problem of output consensus in nonlinear non-minimum phase systems. The main contribution of the paper is to guarantee achieving consensus in the presence of unstable zero dynamics. To achieve this goal, an output redefinition method is proposed. The new outputs of agents are functions of original outputs and internal states and defined such that the dynamics of agents a...
متن کامل