# Spectral-norm risk rates for multi-taper estimation of Gaussian processes

@inproceedings{Romero2021SpectralnormRR, title={Spectral-norm risk rates for multi-taper estimation of Gaussian processes}, author={J. Romero and Michael Speckbacher}, year={2021} }

We consider the estimation of the covariance of a stationary Gaussian process on a multi-dimensional grid from observations taken on a general acquisition domain. We derive spectral-norm risk rates for multi-taper estimators. When applied to one dimensional acquisition intervals, these show that Thomson’s classical multi-taper has optimal risk rates, as they match known benchmarks. We also extend existing lower risk bounds to multidimensional grids and conclude that multi-taper estimators…

## References

SHOWING 1-10 OF 26 REFERENCES

Spectral estimation of irregularly sampled multidimensional processes by generalized prolate spheroidal sequences

- Physics, Computer ScienceIEEE Trans. Acoust. Speech Signal Process.
- 1988

A nonparametric spectral estimation method is presented for bandlimited random processes that have been sampled at arbitrary points in one or more dimensions and avoids several problems typically associated with irregularly sampled data and multidimensional processes.

Multidimensional multitaper spectral estimation

- Mathematics, Computer ScienceSignal Process.
- 1997

A very simple generalization of the one-dimensional multitaper spectral estimation technique to the multi-dimensional case, and several properties of the novel tapers are listed.

Thomson's Multitaper Method Revisited

- Computer Science, EngineeringArXiv
- 2021

This work revisits Thomson’s multitaper method from a linear algebra perspective involving subspace projections and derives an -approximation to the multitaper spectral estimate which can be evaluated on a grid of frequencies using O(log(NW ) log 1 ) FFTs instead of K = O(NW ), which protects against spectral leakage.

Multitaper estimation on arbitrary domains

- Mathematics, Computer ScienceSIAM J. Imaging Sci.
- 2020

It is shown that the multitaper estimator only depends on the linear space spanned by the tapers, and may be replaced by proxy tapers spanning the same subspace (validating the common practice of using partially converged solutions to the Slepian eigenproblem as tapers).

Optimal rates of convergence for estimating Toeplitz covariance matrices

- Mathematics
- 2013

Toeplitz covariance matrices are used in the analysis of stationary stochastic processes and a wide range of applications including radar imaging, target detection, speech recognition, and…

Spectrum estimation and harmonic analysis

- MathematicsProceedings of the IEEE
- 1982

In the choice of an estimator for the spectrum of a stationary time series from a finite sample of the process, the problems of bias control and consistency, or "smoothing," are dominant. In this…

Hanson-Wright inequality and sub-gaussian concentration

- Mathematics
- 2013

In this expository note, we give a modern proof of Hanson-Wright inequality for quadratic forms in sub-gaussian random variables.We deduce a useful concentration inequality for sub-gaussian random…

Spatiospectral concentration in the Cartesian plane

- Mathematics, Physics
- 2010

We pose and solve the analogue of Slepian’s time–frequency concentration problem in the two-dimensional plane, for applications in the natural sciences. We determine an orthogonal family of strictly…

A note on the Hanson-Wright inequality for random vectors with dependencies

- Mathematics
- 2014

We prove that quadratic forms in isotropic random vectors $X$ in $\mathbb{R}^n$, possessing the convex concentration property with constant $K$, satisfy the Hanson-Wright inequality with constant…

L∞-Multivariate approximation theory

- Mathematics
- 1969

This paper is concerned with multivariate analogues of
many standard results in the L∞-approximation theory of functions of one real variable, cf. [12], [13], [14], [20], [21] and [24]. In…