ADAPTATION FOR NONPARAMETRIC ESTIMATORS OF LOCALLY STATIONARY PROCESSES
提出了两种自适应带宽选择方法，用于最小化局部平稳过程中非参数估计的均方误差，并推导了渐近性质，适用于包括非线性过程在内的多种统计情形。
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 a method based on contrast minimization and derive asymptotic properties of both methods. The results are applicable for different statistics under a general setting of local stationarity including nonlinear processes. At the same time, we deepen the general framework for local stationarity based on stationary approximations. For example, a general Bernstein inequality is derived for such processes. The properties of the bandwidth selection methods are also investigated in several simulation studies.