Wavelets (see [2, 3, 4]) are a recent mathematical tool that is applied in signal processing, numerical mathematics and statistics. The wavelet transform allows to follow data in the frequency as well as time domain, to compute efficiently the wavelet coefficients using fast algorithm, to separate approximations from details. Due to these properties, the wavelet transform is suitable for analyzing and forecasting in time series. In this paper, Box-Jenkins models (see [1, 5]) combined with wavelets are used to the prediction of a time series behavior. The described method is demonstrated on an example from practice in the conclusion.
@article{702678,
title = {Wavelets and prediction in time series},
booktitle = {Programs and Algorithms of Numerical Mathematics},
series = {GDML\_Books},
publisher = {Institute of Mathematics AS CR},
address = {Prague},
year = {2015},
pages = {156-162},
url = {http://dml.mathdoc.fr/item/702678}
}
Mošová, Vratislava. Wavelets and prediction in time series, dans Programs and Algorithms of Numerical Mathematics, GDML_Books, (2015), pp. 156-162. http://gdmltest.u-ga.fr/item/702678/