An official website of the United States government
Information from frequency bands in biomedical time series provides useful summaries of the observed signal. Many existing methods consider summaries of the time series obtained over a few well-known, pre-defined frequency bands of interest. However, there is a dearth of data-driven methods for identifying frequency bands that optimally summarize frequency-domain information in the time series. A new method to identify partition points in the frequency space of a multivariate locally stationary time series is proposed. These partition points signify changes across frequencies in the time-varying behavior of the signal and provide frequency band summary measures that best preserve nonstationary dynamics of the observed series. An $$L_2$$-norm based discrepancy measure that finds differences in the time-varying spectral density matrix is constructed, and its asymptotic properties are derived. New nonparametric bootstrap tests are also provided to identify significant frequency partition points and to identify components and cross-components of the spectral matrix exhibiting changes over frequencies. Finite-sample performance of the proposed method is illustrated via simulations. The proposed method is used to develop optimal frequency band summary measures for characterizing time-varying behavior in resting-state electroencephalography time series, as well as identifying components and cross-components associated with each frequency partition point.
more »
« less
This content will become publicly available on January 1, 2027
The Dual Frequency Spectral Density Function of Locally Periodic Stationary Processes With an Application to Testing for Correlation Between Different Frequency Bands of a Time Series
ABSTRACT Harmonizable processes are a class of nonstationary time series, that are characterized by their dependence between different frequencies of a time series. The covariance between two frequencies is the dual frequency spectral density, an object analogous to the spectral density function. Local stationarity is another popular form of nonstationarity, though thus far, little attention has been paid to the dual frequency spectral density of a locally stationary process. The focus of this paper is on the dual frequency spectral density of local stationary time series and locally periodic stationary time series, its natural extension. We show that there are some subtle but important differences between the dual frequency spectral density of an almost periodic stationary process and a locally periodic stationary time series. Estimation of the dual frequency spectral density is typically done by smoothing the dual frequency periodogram. We study the sampling properties of this estimator under the assumption of locally periodic stationarity. In particular, we obtain a Gaussian approximation for the smoothed dual frequency periodogram over a group of frequencies, allowing for the number of frequency lags to grow with sample size. These results are used to test for correlation between different frequency bands in the time series. The variance of the smooth dual frequency periodogram is quite complex. However, by identifying which covariances are the most pertinent we propose a nonparametric method for consistently estimating the variance. This is necessary for constructing confidence intervals or testing aspects of the dual frequency spectral density. Simulations are given to illustrate our results.
more »
« less
- Award ID(s):
- 2210726
- PAR ID:
- 10657843
- Publisher / Repository:
- Wiley
- Date Published:
- Journal Name:
- Journal of Time Series Analysis
- Volume:
- 47
- Issue:
- 1
- ISSN:
- 0143-9782
- Page Range / eLocation ID:
- 158 to 173
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract While the Lomb–Scargle periodogram is foundational to astronomy, it has a significant shortcoming: the variance in the estimated power spectrum does not decrease as more data are acquired. Statisticians have a 60 yr history of developing variance-suppressing power spectrum estimators, but most are not used in astronomy because they are formulated for time series with uniform observing cadence and without seasonal or daily gaps. Here we demonstrate how to apply the missing-data multitaper power spectrum estimator to spacecraft data with uniform time intervals between observations but missing data during thruster fires or momentum dumps. TheF-test for harmonic components may be applied to multitaper power spectrum estimates to identify statistically significant oscillations that would not rise above a white noise–based false alarm probability. Multitapering improves the dynamic range of the power spectrum estimate and suppresses spectral window artifacts. We show that the multitaper–F-test combination applied to Kepler observations of KIC 6102338 detects differential rotation without requiring iterative sinusoid fitting and subtraction. Significant signals reside at harmonics of both fundamental rotation frequencies and suggest an antisolar rotation profile. Next we use the missing-data multitaper power spectrum estimator to identify the oscillation modes responsible for the complex “scallop-shell” shape of the K2 light curve of EPIC 203354381. We argue that multitaper power spectrum estimators should be used for all time series with regular observing cadence.more » « less
-
Frequency combs have revolutionized the measurement of time and frequency since their invention in 2000, and have a wide array of applications to applications that range from basic science applications, to a wide array of sensing applications, to commercial applications, to military applications, and the list goes on. Noise poses a fundamental limit to these systems, and calculating its impact play a critical role in system design. Frequency combs are created by modelocked laser systems that emit a periodic train of short pulses. Laser systems are complex nonlinear systems and the usual method for determining the impact of noise is to carry out computationally-expensive Monte Carlo methods. That limits the parameter range over which it is possible to study the noise impact. We have developed a new approach based on dynamical systems methods. In our approach, we determine a stationary state of the laser system as parameters vary solving a root-finding problem [Wang1]. Starting from a stationary state, we determine all the eigenvalues and eigenvalues of the linearized system. The variance of the amplitudes of the eigenvalues obey either random walk of Langevin equations [Menyuk]. Starting from that point, we can determine the power spectral density of the key laser parameters (amplitude jitter, timing jitter, frequency jitter, phase jitter) [Wang2]. We applied this approach to SESAM lasers and found that we were able to reproduce a computation that took 20 minutes on a cluster with 256 cores with a computation that took less than 4 minutes on a desktop computer.more » « less
-
Context.Variability is a ubiquitous feature of active galactic nuclei (AGNs), and the characterisation of this variability is crucial to constraining its physical mechanism and proper applications in AGN studies. The advent of all-sky and high-cadence optical surveys allows more accurate measurements of AGN variability down to short timescales as well as direct comparisons with X-ray variability from the same sample of sources. Aims.We aim to analyse the optical power spectral density (PSD) of AGNs with measured X-ray PSDs. Methods.We used light curves from the All-Sky Automated Survey for SuperNovae (ASAS-SN) and the Transiting Exoplanet Survey Satellite (TESS) and used the Lomb-Scargle periodogram to obtain PSDs. The joint optical PSD is measured over up to six orders of magnitude in frequency space on timescales of minutes to a decade. We fitted either a damped random walk (DRW) or a broken power law (BPL) model to constrain the PSD model and break frequency. Results.We find a set of break frequencies (≲10−2day−1) from DRW and BPL fits that generally confirm previously reported correlations between break frequencies and the black hole mass. In addition, we find a second set of break frequencies at higher frequencies (> 10−2day−1). We observe a potential weak correlation between the high-frequency breaks with the X-ray break frequencies and the black hole mass. We further explored the dependence of the correlations on other AGN parameters, finding that adding X-ray, optical, or bolometric luminosity as the third correlation parameter can substantially improve the correlation significances. The newly identified high-frequency optical breaks can constrain different aspects of the physics of AGNs.more » « less
-
We propose computationally efficient methods for estimating stationary multivariate spatial and spatial–temporal spectra from incomplete gridded data. The methods are iterative and rely on successive imputation of data and updating of model estimates. Imputations are done according to a periodic model on an expanded domain. The periodicity of the imputations is a key feature that reduces edge effects in the periodogram and is facilitated by efficient circulant embedding techniques. In addition, we describe efficient methods for decomposing the estimated cross spectral density function into a linear model of coregionalization plus a residual process. The methods are applied to two storm datasets, one of which is from Hurricane Florence, which struck the southeastern United States in September 2018. The application demonstrates how fitted models from different datasets can be compared, and how the methods are computationally feasible on datasets with more than 200,000 total observations.more » « less
