Empirical spectral processes for locally stationary time series
Dahlhaus, Rainer ; Polonik, Wolfgang
Bernoulli, Tome 15 (2009) no. 1, p. 1-39 / Harvested from Project Euclid
A time-varying empirical spectral process indexed by classes of functions is defined for locally stationary time series. We derive weak convergence in a function space, and prove a maximal exponential inequality and a Glivenko–Cantelli-type convergence result. The results use conditions based on the metric entropy of the index class. In contrast to related earlier work, no Gaussian assumption is made. As applications, quasi-likelihood estimation, goodness-of-fit testing and inference under model misspecification are discussed. In an extended application, uniform rates of convergence are derived for local Whittle estimates of the parameter curves of locally stationary time series models.
Publié le : 2009-02-15
Classification:  asymptotic normality,  empirical spectral process,  locally stationary processes,  non-stationary time series,  quadratic forms
@article{1233669881,
     author = {Dahlhaus, Rainer and Polonik, Wolfgang},
     title = {Empirical spectral processes for locally stationary time series},
     journal = {Bernoulli},
     volume = {15},
     number = {1},
     year = {2009},
     pages = { 1-39},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1233669881}
}
Dahlhaus, Rainer; Polonik, Wolfgang. Empirical spectral processes for locally stationary time series. Bernoulli, Tome 15 (2009) no. 1, pp.  1-39. http://gdmltest.u-ga.fr/item/1233669881/