ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 1710.09072
26
30

Asymptotically Efficient Estimation of Smooth Functionals of Covariance Operators

25 October 2017
V. Koltchinskii
ArXivPDFHTML
Abstract

Let XXX be a centered Gaussian random variable in a separable Hilbert space H{\mathbb H}H with covariance operator Σ.\Sigma.Σ. We study a problem of estimation of a smooth functional of Σ\SigmaΣ based on a sample X1,…,XnX_1,\dots ,X_nX1​,…,Xn​ of nnn independent observations of X.X.X. More specifically, we are interested in functionals of the form ⟨f(Σ),B⟩,\langle f(\Sigma), B\rangle,⟨f(Σ),B⟩, where f:R↦Rf:{\mathbb R}\mapsto {\mathbb R}f:R↦R is a smooth function and BBB is a nuclear operator in H.{\mathbb H}.H. We prove concentration and normal approximation bounds for plug-in estimator ⟨f(Σ^),B⟩,\langle f(\hat \Sigma),B\rangle,⟨f(Σ^),B⟩, Σ^:=n−1∑j=1nXj⊗Xj\hat \Sigma:=n^{-1}\sum_{j=1}^n X_j\otimes X_jΣ^:=n−1∑j=1n​Xj​⊗Xj​ being the sample covariance based on X1,…,Xn.X_1,\dots, X_n.X1​,…,Xn​. These bounds show that ⟨f(Σ^),B⟩\langle f(\hat \Sigma),B\rangle⟨f(Σ^),B⟩ is an asymptotically normal estimator of its expectation EΣ⟨f(Σ^),B⟩{\mathbb E}_{\Sigma} \langle f(\hat \Sigma),B\rangleEΣ​⟨f(Σ^),B⟩ (rather than of parameter of interest ⟨f(Σ),B⟩\langle f(\Sigma),B\rangle⟨f(Σ),B⟩) with a parametric convergence rate O(n−1/2)O(n^{-1/2})O(n−1/2) provided that the effective rank r(Σ):=tr(Σ)∥Σ∥{\bf r}(\Sigma):= \frac{{\bf tr}(\Sigma)}{\|\Sigma\|}r(Σ):=∥Σ∥tr(Σ)​ (tr(Σ){\rm tr}(\Sigma)tr(Σ) being the trace and ∥Σ∥\|\Sigma\|∥Σ∥ being the operator norm of Σ\SigmaΣ) satisfies the assumption r(Σ)=o(n).{\bf r}(\Sigma)=o(n).r(Σ)=o(n). At the same time, we show that the bias of this estimator is typically as large as r(Σ)n\frac{{\bf r}(\Sigma)}{n}nr(Σ)​ (which is larger than n−1/2n^{-1/2}n−1/2 if r(Σ)≥n1/2{\bf r}(\Sigma)\geq n^{1/2}r(Σ)≥n1/2). In the case when H{\mathbb H}H is finite-dimensional space of dimension d=o(n),d=o(n),d=o(n), we develop a method of bias reduction and construct an estimator ⟨h(Σ^),B⟩\langle h(\hat \Sigma),B\rangle⟨h(Σ^),B⟩ of ⟨f(Σ),B⟩\langle f(\Sigma),B\rangle⟨f(Σ),B⟩ that is asymptotically normal with convergence rate O(n−1/2).O(n^{-1/2}).O(n−1/2). Moreover, we study asymptotic properties of the risk of this estimator and prove minimax lower bounds for arbitrary estimators showing the asymptotic efficiency of ⟨h(Σ^),B⟩\langle h(\hat \Sigma),B\rangle⟨h(Σ^),B⟩ in a semi-parametric sense.

View on arXiv
Comments on this paper