Il giorno Giovedì 9 Aprile 2015, alle ore 14:30 presso la sede di Prometeia (sala grande, primo piano) via G.Marconi 43, Bologna
Roy CERQUETI (Università di Macerata)
terrà un seminario dal titolo
"Markov chains approximation for resampling and bootstrapping"
Abstract Markov chain theory is proving to be a powerful approach to resample highly nonlinear time series in the context of bootstrap procedures. In this talk a discussion on this theme is provided, with a special focus on the estimation of the memory of a Markov chain (i.e. its order) and on the identification of its relevant states. In particular, the choice of memory lags and the aggregation of irrelevant states are obtained by looking for regularities in the transition probabilities.
At this aim, suitable optimization models are developed under an information theory perspective. More specifically, two competing objectives that a researcher will in general pursue when dealing with bootstrapping are considered: preserving the “structural” similarity between the original and the simulated series and assuring a controlled diversification of the latter. Several critical issues emerge and are analyzed, such as the remarkable complexity of the optimization problem due to the high cardinality of the admissible regions and the presence of interconnections between more than one resampled series in a multivariate framework. Numerical experiments based on real data validate the theoretical treatment of the problem.
--- Questa e-mail è priva di virus e malware perché è attiva la protezione avast! Antivirus. http://www.avast.com