Differentiability of M-functionals of Location and Scatter Based on T Likelihoods
نویسنده
چکیده
The paper aims at finding widely and smoothly defined nonparametric location and scatter functionals. As a convenient vehicle, maximum likelihood estimation of the location vector μ and scatter matrix Σ of an elliptically symmetric t distribution on R with degrees of freedom ν > 1 extends to an M-functional defined on all probability distributions P in a weakly open, weakly dense domain U . Here U consists of P putting not too much mass in hyperplanes of dimension < d, as shown for empirical measures by Kent and Tyler (Ann. Statist. 1991). It is shown here that (μ,Σ) is analytic on U , for the bounded Lipschitz norm, or for d = 1, for the sup norm on distribution functions. For k = 1, 2, ..., and other norms, depending on k and more directly adapted to t functionals, one has continuous differentiability of order k, allowing the delta-method to be applied to (μ,Σ) for any P in U , which can be arbitrarily heavy-tailed. These results imply asymptotic normality of the correspondingM-estimators (μn,Σn). In dimension d = 1 only, the tν functional (μ, σ) extends to be defined and weakly continuous at all P .
منابع مشابه
Differentiability of t-functionals of location and scatter
The paper aims at finding widely and smoothly defined nonparametric location and scatter functionals. As a convenient vehicle, maximum likelihood estimation of the location vector μ and scatter matrix Σ of an elliptically symmetric t distribution on R with degrees of freedom ν > 1 extends to an M-functional defined on all probability distributions P in a weakly open, weakly dense domain U . Her...
متن کاملSome facts about functionals of location and scatter
Assumptions on a likelihood function, including a local GlivenkoCantelli condition, imply the existence of M-estimators converging to an Mfunctional. Scatter matrix-valued estimators, defined on all empirical measures on R for d ≥ 2, and equivariant under all, including singular, affine transformations, are shown to be constants times the sample covariance matrix. So, if weakly continuous, they...
متن کاملOn the Breakdown Properties of Some Multivariate M-Functionals
For probability distributions on IR, a detailed study of the breakdown properties of some multivariate M-functionals related to Tyler’s (1987a) “distribution-free” M-functional of scatter is given. These include a symmetrized version of Tyler’s M-functional of scatter, and the multivariate t M-functionals of location and scatter. It is shown that for “smooth” distributions, the (contamination) ...
متن کاملTukey Depth-based Multivariate Trimmed Means
We investigate the asymptotic behavior of two types of Tukey depth-based multivariate trimmed means. Sufficient conditions for asymptotic normality of these location estimators are given. Two approaches to trimming are distinguished and central limit theorems are derived for each one. Asymptotic normality is proved using Hadamard differentiability of the location functionals. In the one-dimensi...
متن کامل