國立高雄大學圖資館 |

Language: English

Back

Multiscale forecasting models

Maggi, Lida Mercedes Barba.

Multiscale forecasting models

Record Type:	Electronic resources : Monograph/item
Title/Author:	Multiscale forecasting modelsby Lida Mercedes Barba Maggi.
Author:	Maggi, Lida Mercedes Barba.
Published:	Cham :Springer International Publishing :2018.
Description:	xxiv, 124 p. :ill., digital ;24 cm.
Contained By:	Springer eBooks
Subject:	Time-series analysis.
Online resource:	http://dx.doi.org/10.1007/978-3-319-94992-5
ISBN:	9783319949925$q(electronic bk.)

Multiscale forecasting models
Maggi, Lida Mercedes Barba.

Multiscale forecasting models[electronic resource] /by Lida Mercedes Barba Maggi. - Cham :Springer International Publishing :2018. - xxiv, 124 p. :ill., digital ;24 cm.

Dedication -- Foreword -- Preface -- Acknowledgement -- List of Tables -- List of Figures -- Acronyms -- 1. Times Series Analysis -- 2. Forecasting based on Hankel Singular Value Decomposition -- 3.Multi-step ahead forecasting -- 4. Multilevel Singular Value Decomposition.

This book presents two new decomposition methods to decompose a time series in intrinsic components of low and high frequencies. The methods are based on Singular Value Decomposition (SVD) of a Hankel matrix (HSVD) The proposed decomposition is used to improve the accuracy of linear and nonlinear auto-regressive models. Linear Auto-regressive models (AR, ARMA and ARIMA) and Auto-regressive Neural Networks (ANNs) have been found insufficient because of the highly complicated nature of some time series. Hybrid models are a recent solution to deal with non-stationary processes which combine pre-processing techniques with conventional forecasters, some pre-processing techniques broadly implemented are Singular Spectrum Analysis (SSA) and Stationary Wavelet Transform (SWT) Although the flexibility of SSA and SWT allows their usage in a wide range of forecast problems, there is a lack of standard methods to select their parameters. The proposed decomposition HSVD and Multilevel SVD are described in detail through time series coming from the transport and fishery sectors. Further, for comparison purposes, it is evaluated the forecast accuracy reached by SSA and SWT, both jointly with AR-based models and ANNs.

ISBN: 9783319949925$q(electronic bk.)

Standard No.: 10.1007/978-3-319-94992-5doiSubjects--Topical Terms:

181890
Time-series analysis.

LC Class. No.: QA280

Dewey Class. No.: 519.55

Multiscale forecasting models
LDR:02432nmm a2200325 a 4500 001 543659
003 DE-He213
005 20180823141935.0
006 m d
007 cr nn 008maaau
008 190430s2018 gw s 0 eng d
020 $a 9783319949925$q(electronic bk.)
020 $a 9783319949918$q(paper)
024 7 $a 10.1007/978-3-319-94992-5 $2 doi
035 $a 978-3-319-94992-5
040 $a GP $c GP
041 0 $a eng
050 4 $a QA280
072 7 $a UYQ $2 bicssc
072 7 $a TJFM1 $2 bicssc
072 7 $a COM004000 $2 bisacsh
082 0 4 $a 519.55 $2 23
090 $a QA280 $b .M193 2018
100 1 $a Maggi, Lida Mercedes Barba. $3 821949
245 1 0 $a Multiscale forecasting models $h [electronic resource] / $c by Lida Mercedes Barba Maggi.
260 $a Cham : $b Springer International Publishing : $b Imprint: Springer, $c 2018.
300 $a xxiv, 124 p. : $b ill., digital ; $c 24 cm.
505 0 $a Dedication -- Foreword -- Preface -- Acknowledgement -- List of Tables -- List of Figures -- Acronyms -- 1. Times Series Analysis -- 2. Forecasting based on Hankel Singular Value Decomposition -- 3.Multi-step ahead forecasting -- 4. Multilevel Singular Value Decomposition.
520 $a This book presents two new decomposition methods to decompose a time series in intrinsic components of low and high frequencies. The methods are based on Singular Value Decomposition (SVD) of a Hankel matrix (HSVD) The proposed decomposition is used to improve the accuracy of linear and nonlinear auto-regressive models. Linear Auto-regressive models (AR, ARMA and ARIMA) and Auto-regressive Neural Networks (ANNs) have been found insufficient because of the highly complicated nature of some time series. Hybrid models are a recent solution to deal with non-stationary processes which combine pre-processing techniques with conventional forecasters, some pre-processing techniques broadly implemented are Singular Spectrum Analysis (SSA) and Stationary Wavelet Transform (SWT) Although the flexibility of SSA and SWT allows their usage in a wide range of forecast problems, there is a lack of standard methods to select their parameters. The proposed decomposition HSVD and Multilevel SVD are described in detail through time series coming from the transport and fishery sectors. Further, for comparison purposes, it is evaluated the forecast accuracy reached by SSA and SWT, both jointly with AR-based models and ANNs.
650 0 $a Time-series analysis. $3 181890
650 0 $a Forecasting $x Mathematical models. $3 206969
650 1 4 $a Computer Science. $3 212513
650 2 4 $a Artificial Intelligence (incl. Robotics) $3 252959
650 2 4 $a Probability and Statistics in Computer Science. $3 274053
650 2 4 $a Algebra. $3 188312
710 2 $a SpringerLink (Online service) $3 273601
773 0 $t Springer eBooks
856 4 0 $u http://dx.doi.org/10.1007/978-3-319-94992-5
950 $a Computer Science (Springer-11645)