Spectral Models for Locally Stationary Processes (Joint with Tommaso Proietti and Stefano Grassi)

Abstract:

A class of models for the time-varying spectrum of a locally stationary process is introduced. The models are specified in the frequency domain and the class depends on a power parameter that applies to the spectrum so that it can be locally represented by a finite Fourier polynomial. The coefficients of the polynomial have an interpretation as generalized autocovariances whose smooth time variation is determined according to a linear combination of logistic transition functions of the time index. Estimation is carried out in the frequency domain based on the generalized Whittle likelihood. Parametric inference is developed.