Subsampling for nonstationary time series with non-zero mean function

Anna Dudek (1, 2) , Łukasz Lenart (3)

1 UR2 - Université de Rennes 2
2 AGH UST - AGH University of Science and Technology [Krakow, PL]
3 Cracow University of Economics

In this paper a subsampling approach for nonstationary time series with a non-zero mean function is proposed. It is applied for periodically and almost periodically processes. Two statistical tests are constructed. An example with real data is presented.

Mots clés: almost periodically correlated; autocovariance function; Fourier coefficients; periodicity; testing

Domaines: Statistiques [stat]

Anna Dudek, Łukasz Lenart. Subsampling for nonstationary time series with non-zero mean function. 2017. ⟨hal-01415632v3⟩