ADAPTATION FOR NONPARAMETRIC ESTIMATORS OF LOCALLY STATIONARY PROCESSES

Two adaptive bandwidth selection methods for minimizing the mean squared error of nonparametric estimators in locally stationary processes are proposed. We investigate a cross-validation approach and method based on contrast minimization derive asymptotic properties both methods. The results applicable different statistics under general setting local stationarity including nonlinear processes. At same time, we deepen framework approximations. For example, Bernstein inequality is derived such also investigated several simulation studies.