A nonparametric test for stationarity in functional time series

TL;DR

This paper introduces a new nonparametric frequency domain test for stationarity in functional time series, utilizing spectral density operators and Hilbert-Schmidt inner products, with demonstrated good finite sample performance.

Contribution

The paper develops a novel spectral density-based measure for stationarity in functional time series and provides an asymptotic normality result for the estimator, enabling a simple frequency domain test.

Findings

The test accurately detects stationarity in simulations.

The method performs well with finite samples.

Application to temperature data demonstrates practical utility.

Abstract

We propose a new measure for stationarity of a functional time series, which is based on an explicit representation of the $L^2$-distance between the spectral density operator of a non-stationary process and its best ($L^2$-)approximation by a spectral density operator corresponding to a stationary process. This distance can easily be estimated by sums of Hilbert-Schmidt inner products of periodogram operators (evaluated at different frequencies), and asymptotic normality of an appropriately standardized version of the estimator can be established for the corresponding estimate under the null hypothesis and alternative. As a result we obtain a simple asymptotic frequency domain level $\alpha$ test (using the quantiles of the normal distribution) for the hypothesis of stationarity of functional time series. Other applications such as asymptotic confidence intervals for a measure of stationarity or the construction of tests for "relevant deviations from stationarity", are also briefly mentioned. We demonstrate in a small simulation study that the new method has very good finite sample properties. Moreover, we apply our test to annual temperature curves.