Empirical spectral processes for stationary state space models

TL;DR

This paper develops a central limit theorem for spectral processes in multivariate continuous-time state space models, enabling better inference and goodness-of-fit testing for such models.

Contribution

It introduces weaker assumptions for the asymptotic distribution of spectral estimators in continuous-time state space models, with explicit covariance representations.

Findings

Derived a CLT for the normalized weighted integrated periodogram.

Established weak convergence to a Gaussian process.

Provided explicit limit distributions for goodness-of-fit tests.

Abstract

In this paper, we consider function-indexed normalized weighted integrated periodograms for equidistantly sampled multivariate continuous-time state space models which are multivariate continuous-time ARMA processes. Thereby, the sampling distance is fixed and the driving L\'evy process has at least a finite fourth moment. Under different assumptions on the function space and the moments of the driving L\'evy process we derive a central limit theorem for the function-indexed normalized weighted integrated periodogram. Either the assumption on the function space or the assumption on the existence of moments of the L\'evy process is weaker. Furthermore, we show the weak convergence in both the space of continuous functions and in the dual space to a Gaussian process and give an explicit representation of the covariance function. The results can be used to derive the asymptotic behavior of the Whittle estimator and to construct goodness-of-fit test statistics as the Grenander-Rosenblatt statistic and the Cram\'er-von Mises statistic. We present the exact limit distributions of both statistics and show their performance through a simulation study.