Estimation de la densité spectrale d'un processus en temps continu par échantillonage aléatoire

Abstract

In this thesis we study the estimator of the spectral density of a continuous-time processes, with random sampling, wich be proposed by Lii and Masry in 1994. Next we consider the case of an unknown mean process, we show that the bias of the estimator is asymptotically null. Finally a simulation work permits us to complete the theoretical study.