Mathematical Framework for Pseudo-Spectra of Linear Stochastic Difference Equations

Abstract

Although spectral analysis of stationary stochastic processes has solid mathematical foundations, this is not always so for some non-stationary cases. Here, we establish a rigorous mathematical extension of the classic Fourier spectrum to the case in which there are AR roots in the unit circle, ie, the transfer function of the linear time-invariant filter has poles on the unit circle. To achieve it we: embed the classical problem in a wider framework, extend the Discrete Time Fourier Transform and defined a new Extended Fourier Transform pair pseudo-covariance function/pseudo-spectrum. Our approach is a proper extension of the classical spectral analysis, within which the Fourier Transform pair auto-covariance function/spectrum is a particular case. Consequently spectrum and pseudo-spectrum coincide when the first one is defined.Keywords Spectral analysis, time series, non-stationarity, frequency domain, pseudocovariance function, linear stochastic difference equations, partial inner product, Extended Fourier Transform. JL Classification C00, C22This working paper has been accepted for publication in a future issue of IEEE Transactions on Signal Processing. Content may change prior to final publication. Abstract-Although spectral analysis of stationary stochastic processes has solid mathematical foundations, this is not always so for some non-stationary cases. Here, we establish a rigorous mathematical extension of the classic Fourier spectrum to the case in which there are AR roots in the unit circle, ie, the transfer function of the linear time-invariant filter has poles on the unit circle. To achieve it we: embed the classical problem in a wider framework, extend the Discrete Time Fourier Transform and defined a new Extended Fourier Transform pair pseudocovariance function/pseudo-spectrum. Our approach is a proper extension of the classical spectral analysis, within which the Fourier Transform pair auto-covariance function/spectrum is a particular case. Consequently spectrum and pseudo-spectrum coincide when the first one is defined.

Casting vector time series: algorithms for forecasting, imputation, and signal extraction

“…One seeks a WK filter Ψ(B) such that the mean squared error (MSE) of estimating s t via Ψ(B)x t is minimized. The solution is given in [2], and extended to the multivariate case in [27]: essentially, Ψ(e −iλ ) is given by a ratio of pseudospectra [4] for signal to data process. Hence, if a model with latent components has been fitted the spectral densities for signal and noise can be calculated, and the coefficients ψ j computed.…”

Evaluating early warning and coincident indicators of business cycles using smooth trends

“…The LDHR is a linear estimation procedure that also provides an automatic identification of the complete DHR model. It exploits the algebraic structure of the pseudo-spectrum functions (see Bujosa, Bujosa, & García-Ferrer, 2015) to avoid the poles associated with the unit modulus AR roots of the pseudo-spectrum of DHR models f dhr (𝜔).…”

Difference equation−based transient-state and steady-state analysis of flyback converter circuit

“…Difference equations (Brand, 1966;Miller, 1968;Bainov and Stamova, 1997) allow a nonlinear system to transform into a linear one. Therefore, this technique is used in many applications of electrical engineering from signal processing to power system tools (Shoihet and Slonim, 2010;Babiarz, 2015;Li et al, 2017;Ngoc, 2018;Shen et al, 2016;Bujosa and Ferrer, 2015;Mirkovic et al, 2014). The nonlinear converter topologies (Flyback, Forward, Push-Pull, halfbridge and full-bridge topologies) are solved by using genetic algorithms in Versele et al (2014).…”