Prediction of time series by statistical learning: general losses and fast rates
|Title:||Prediction of time series by statistical learning: general losses and fast rates||Authors:||Alquier, Pierre
|Permanent link:||http://hdl.handle.net/10197/8374||Date:||31-Dec-2013||Abstract:||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 , where n is the sample size and d the dimension of the set of predictors. Under the same weak dependence conditions, we extend this result to any convex Lipschitz loss function. We also identify a condition on the parameter space that ensures similar rates for the classical penalized ERM procedure. We apply this method for quantile forecasting of the French GDP. Under additional conditions on the loss functions (satisfied by the quadratic loss function) and for uniformly mixing processes, we prove that the Gibbs estimator actually achieves fast rates of convergence d/n. We discuss the optimality of these different rates pointing out references to lower bounds when they are available. In particular, these results bring a generalization the results of  on sparse regression estimation to some autoregression.||Type of material:||Journal Article||Publisher:||De Gruyter||Copyright (published version):||2013 the Authors||Keywords:||Recommender systems;Statistical learning theory;Time series forecasting;PAC-Bayesian bounds;Weak dependence;Mixing;Oracle inequalities;Fast rates;GDP forecasting||DOI:||10.2478/demo-2013-0004||Language:||en||Status of Item:||Peer reviewed|
|Appears in Collections:||Mathematics and Statistics Research Collection|
Insight Research Collection
Show full item record
This item is available under the Attribution-NonCommercial-NoDerivs 3.0 Ireland. No item may be reproduced for commercial purposes. For other possible restrictions on use please refer to the publisher's URL where this is made available, or to notes contained in the item itself. Other terms may apply.