Reconciling the Gaussian and Whittle likelihood with an application to estimation in the frequency domain
From MaRDI portal
Publication:2054529
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.
Recommendations
Cites work
- scientific article; zbMATH DE number 3852259 (Why is no real title available?)
- scientific article; zbMATH DE number 3854249 (Why is no real title available?)
- scientific article; zbMATH DE number 3806621 (Why is no real title available?)
- scientific article; zbMATH DE number 3263751 (Why is no real title available?)
- scientific article; zbMATH DE number 3070807 (Why is no real title available?)
- scientific article; zbMATH DE number 3086018 (Why is no real title available?)
- A necessary and sufficient condition for asymptotic independence of discrete Fourier transforms under short- and long-range dependence
- A note on Herglotz's theorem for time series on function spaces
- APPROXIMATE CONFIDENCE INTERVALS
- An Asymptotic Result for the Finite Predictor.
- An approximate inverse for the covariance matrix of moving average and autoregressive processes
- An efficient taper for potentially overdifferenced long-memory time series
- Approximations for stationary covariance matrices and their inverses with application to ARMA models
- Asymptotic properties of least-squares estimates of parameters of the spectrum of a stationary non-deterministic time-series
- Asymptotically efficient autoregressive model selection for multistep prediction
- Baxter's inequality and convergence of finite predictors of multivariate stochastic processes
- Baxter's inequality for finite predictor coefficients of multivariate long-memory stationary processes
- Bayesian Estimation of the Spectral Density of a Time Series
- Beyond Whittle: nonparametric correction of a parametric likelihood with a focus on Bayesian time series analysis
- Duals of random vectors and processes with applications to prediction problems with missing values
- Edge effects and efficient parameter estimation for stationary random fields
- Explicit representation of finite predictor coefficients and its applications
- Fourier analysis of stationary time series in function space
- Gaussian semiparametric estimation of long range dependence
- Large sample inference for long memory processes
- Large-sample properties of parameter estimates for strongly dependent stationary Gaussian time series
- Local Whittle estimation of fractional integration for nonlinear processes
- Nonparametric high resolution spectral estimation
- Nonstationarity-extended local Whittle estimation
- On the Inversion of the Sample Covariance Matrix in a Stationary Autoregressive Process
- On the inverses of some patterned matrices arising in the theory of stationary time series
- On the range of validity of the autoregressive sieve bootstrap
- Order selection for same-realization predictions in autoregressive processes
- Orthogonal samples for estimators in time series
- Predictive, finite-sample model choice for time series under stationarity and non-stationarity
- REDUCTION OF THE ASYMPTOTIC BIAS OF AUTOREGRESSIVE AND SPECTRAL ESTIMATORS BY TAPERING
- SPECTRAL ANALYSIS WITH TAPERED DATA
- Small sample effects in time series analysis: A new asymptotic theory and a new estimate
- Spectral methods for small sample time series: A complete periodogram approach
- The debiased Whittle likelihood
- The evaluation of certain quadratic forms occurring in autoregressive model fitting
- Time series. Data analysis and theory.
- WHITTLE ESTIMATION OF ARCH MODELS
Cited in
(14)- A nonparametrically corrected likelihood for Bayesian spectral analysis of multivariate time series
- Contiguity of the Whittle measure for a Gaussian time series
- Inference with the Whittle Likelihood: A Tractable Approach Using Estimating Functions
- Explicit formulas for the inverses of Toeplitz matrices, with applications
- On the asymptotic behavior of a finite section of the optimal causal filter
- scientific article; zbMATH DE number 4186937 (Why is no real title available?)
- Statistical analysis of irregularly spaced spatial data in frequency domain
- On the equivalence of time and frequency domain maximum likelihood estimation
- Frequency domain generalized empirical likelihood method
- A prediction perspective on the Wiener–Hopf equations for time series
- Spectral methods for small sample time series: A complete periodogram approach
- Higher‐order asymptotics of minimax estimators for time series
- A Note on Whittle's Likelihood
- Posterior consistency for the spectral density of non‐Gaussian stationary time series
This page was built for publication: Reconciling the Gaussian and Whittle likelihood with an application to estimation in the frequency domain
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2054529)
