We propose an algorithm to estimate the common density of a stationary process . We suppose that the process is either or -mixing. We provide a model selection procedure based on a generalization of Mallows' and we prove oracle inequalities for the selected estimator under a few prior assumptions on the collection of models and on the mixing coefficients. We prove that our estimator is adaptive over a class of Besov spaces, namely, we prove that it achieves the same rates of convergence as in the i.i.d framework.
View on arXiv