Pierre, Alquier; Xiaoyin, Li; Olivier, Wintenberger - In: Dependence Modeling 1 (2014) January, pp. 65-93
We establish rates of convergences in statistical learning for time series forecasting. Using the PAC-Bayesian approach, slow rates of convergence √ d/n for the Gibbs estimator under the absolute loss were given in a previous work [7], where n is the sample size and d the dimension of the set...