Precise rates in the law of the logarithm for the moment convergence in Hilbert spaces

Ke Ang Fu,Li Xin Zhang

DOI: https://doi.org/10.1007/s10114-008-6529-1

2009-01-01

Abstract:Let X, X n ; n ≥ 1 be a sequence of i.i.d. random variables taking values in a real separable Hilbert space ( H , ‖ · ‖) with covariance operator Σ. Set S n = X 1 + X 2 + ... + X n , n ≥ 1. We prove that, for b > −1, lim_ε↘ 0ε ^2(b + 1)∑_n = 1^∞(logn)^b /n^3 / . - 2 E{S_n - σε√(nlogn)} _ + = σ ^ - 2(b + 1)/^(2b + 3)(b + 1)E Y ^2b + 3 holds if E X = 0, and E‖ X ‖ 2 (log ‖ X ‖) 3 b ∨( b +4) < ∞, where Y is a Gaussian random variable taking value in a real separable Hilbert space with mean zero and covariance operator Σ, and σ 2 denotes the largest eigenvalue of Σ.