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.

Whittle likelihood estimation in INAR(p) process

Whittle likelihood estimation in INAR(1) process

The Invariance Principle for Fractionally Integrated Processes with Strong Near-Epoch Dependent Innovations

A maximum likelihood and regenerative bootstrap approach for estimation and forecasting of INAR(p) processes with zero-inflated innovations

Nonparametric Estimation of Quantiles for A Class of Stationary Processes

Asymptotic behavior of unstable INAR(p) processes

On strongly dependent zero-inflated INAR(1) processes

A new estimation for INAR(1) process with Poisson distribution

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

Stationary underdispersed INAR(1) models based on the backward approach

Change point detection in INAR(p) models via likelihood ratio scanning method

A new bivariate INAR(1) model with paired Poisson-weighted exponential distributed innovations

Comparison of estimation and prediction methods for a zero-inflated geometric INAR(1) process with random coefficients

LOCAL WHITTLE ESTIMATION OF FRACTIONAL INTEGRATION FOR NONLINEAR PROCESSES

On the discrete analogue of the Teissier distribution and its associated INAR(1) process

Poisson Extended Exponential Distribution with Associated INAR(1) Process and Applications

Inferential properties with a novel two parameter Poisson generalized Lindley distribution with regression and application to INAR(1) process

INAR(1) process with weighted negative binomial Lindley distributed innovations and applications to criminal and COVID-19 data

A new INAR model based on Poisson-BE2 innovations

On the novel geometric and negative binomial INAR(1) processes

Semi-parametric goodness-of-fit testing for INAR models