On robustness of model-based bootstrap schemes in nonparametric time series analysis

摘要: Theory in time series analysis is often developed the context of finite-dimensional models for data generating process. Whereas corresponding estimators such as those a conditional mean function are reasonable even if true dependence mechanism more complex structure, it usually necessary to capture whole structure asymptotically bootstrap be valid. However, certain model-based methods remain valid some interesting quantities arising nonparametric statistics. We generalize well-known whitening by windowing principle joint distributions autoregression function. As consequence, we obtain that schemes supremum-type functionals long they mimic consistently. an example, investigate finite order Markov chain general stationary