Abstract:In time series analysis there is an apparent dichotomy between time and frequency domain methods. The aim of this paper is to draw connections between frequency and time domain methods. Our focus will be on reconciling the Gaussian likelihood and the Whittle likelihood. We derive an exact, interpretable, bound between the Gaussian and Whittle likelihood of a second order stationary time series. The derivation is based on obtaining the transformation which is biorthogonal to the discrete Fourier transform of the time series. Such a transformation yields a new decomposition for the inverse of a Toeplitz matrix and enables the representation of the Gaussian likelihood within the frequency domain. We show that the difference between the Gaussian and Whittle likelihood is due to the omission of the best linear predictions outside the domain of observation in the periodogram associated with the Whittle likelihood. Based on this result, we obtain an approximation for the difference between the Gaussian and Whittle likelihoods in terms of the best fitting, finite order autoregressive parameters. These approximations are used to define two new frequency domain quasi-likelihoods criteria. We show that these new criteria can yield a better approximation of the spectral divergence criterion, as compared to both the Gaussian and Whittle likelihoods. In simulations, we show that the proposed estimators have satisfactory finite sample properties.

Frequency-domain correlation: an asymptotically optimum approximation of quadratic likelihood ratio detectors

Evaluation of Bistable Systems for Binary Signal Detection in Symmetric Non-Gaussian Noise

Asymptotically Optimal One-Bit Quantizer Design for Weak-Signal Detection in Generalized Gaussian Noise and Lossy Binary Communication Channel.

Polynomial-approximation-based Locally Optimum Detector for Signals with Symmetric Alpha Stable Noise.

Log Likelihood Ratio Test for Spectrum Sensing With Truncated Covariance Matrix

Optimal Signals and Detectors Based on Correlation and Energy

Adaptive Detection and Estimation for an Unknown Occurring Interval Signal in Correlated Gaussian Noise

Quasi-convexity and Optimal Binary Fusion for Distributed Detection with Identical Sensors in Generalized Gaussian Noise

A Hybrid Autocorrelation-Based Single Frequency Estimator

Nonlinear Processing for Correlation Detection in Symmetric Alpha-Stable Noise.

Reconciling the Gaussian and Whittle Likelihood with an application to estimation in the frequency domain

TESTING FOR SERIAL CORRELATION OF UNKNOWN FORM USING WAVELET METHODS

On Minimax Robust Detection of Stationary Gaussian Signals in White Gaussian Noise

OPTIMAL CORRELATION ESTIMATORS FOR QUANTIZED SIGNALS

Detection with Target-Incurring Orthogonal Subspace Interference

The Detection and Joint Parameters Estimation of LFM Signals Based on a Hybrid Correlation Algorithm on Fractional Frequency Domain

A Likelihood Ratio-Based Detector for QTMS Radar and Noise Radar

A Modified Energy Detector for Random Signals in Gaussian Noise.

Persymmetric Adaptive Detection with Reduced-Dimension Approach

Adaptive GLR-, Rao- and Wald-based CFAR Detectors for a Subspace Signal Embedded in Structured Gaussian Interference

On Minimax Detection of Gaussian Stochastic Sequences and Gaussian Stationary Signals