Publication detail

RE-SAMPLING ESTIMATION IN TIME SERIES

Authors: Kubanová Jana | Linda Bohdan

Year: 2007

Type of publication: článek ve sborníku

Name of source: ISBIS 2007

Publisher name: University of Azores

Place: Azores, Portugal

Page from-to: 145

Titles:

Language	Name	Abstract	Keywords
cze	RE-SAMPLING ESTIMATION IN TIME SERIES	V článku jsou popsány dvě metody analýzy časových řad bootstrapovou metodou,autoregresní model prvního řádu a model klouzavých bloků.Obě metody jsou demonstrovány i na praktických příkladech.Výsledky obou metod jsou v závěru článku porovnány.	časové řady, autoregresní model prvního řádu,klouzavé bloky,bootstrap
eng	RE-SAMPLING ESTIMATION IN TIME SERIES	Many probabilistic models are used to analyse the time series. Two different methods for bootstrapping time series are described in the paper. The simplest model ? a first order autoregressive scheme is described and presented in the practical example. The principle of this method is fitting a model and then sampling from the residuals. The different presented method is called the moving block bootstrap. An important advantage of the method is that is less ?model dependent? than the bootstrapping of residuals approach. The method is dependent on the model that is fit to the original time series. Results of both methods are compared in the paper. The methods for bootstrapping analogous to the solution for the regression models can be used and they are mentioned in the end.	time series, first order autoregressive scheme,moving block bootstrap

Search