Convergence Rate Analysis in Limit Theorems for Nonlinear Functionals of the Second Wiener Chaos

Abstract

The purpose of this paper is to analyze the distribution distance between random vectors derived from the magnitude of the analytic wavelet transform of the squared envelopes of Gaussian processes and their large-scale limits. When the Hurst index of the underlying Gaussian process falls below 1/2, the large-scale limit takes the form of a combination of the second and fourth Wiener chaos. In this case, we employ the non-Stein approach to determine the convergence rate in the Kolmogorov metric. In scenarios where the Hurst index of the underlying Gaussian process exceeds 1/2, the large-scale limit is a correlated chi-distributed random process with two degrees of freedom. The convergence rate in the Wasserstein metric is derived through the application of Malliavin calculus and multidimensional Stein's method. Although the large-scale limits are the same for Hurst indices within the interval (1/2,1), notable differences in the upper bounds of convergence rates are observed among Hurst indices falling within the subintervals (1/2,3/4) and (3/4,1).

43 pages, 3 figures

References in corpus (8)

• On the Analytic Wavelet Transform
• Intermittent process analysis with scattering moments
• A wavelet analysis of the Rosenblatt process: chaos expansion and estimation of the self-similarity parameter
• Limit theorems for weighted nonlinear transformations of Gaussian stationary processes with singular spectra
• Stein's method for normal approximation in Wasserstein distances with application to the multivariate Central Limit Theorem
• Sojourn measures of Student and Fisher-Snedecor random fields
• A deep look into the Dagum family of isotropic covariance functions
• Linear Time Complexity Deep Fourier Scattering Network and Extension to Nonlinear Invariants