Multiscale Forecasting Models

Preface

1.       Time Series and Forecasting

1.2.  Time series

1.3.  Linear Autoregressive Models

1.5.  Hybrid models

1.5.1.         Singular Spectrum Analysis

1.5.2.         Wavelet Transform

1.6.  Forecasting Accuracy Measures

1.7.  Empirical Applications

1.7.1.         Traffic Accidents Forecasting based on AR, ANNs and Hybrid models.

1.7.2.         Anchovy Stock Forecasting based on AR, ANNs and Hybrid models.

1.7.3.         Sardine Stock Forecasting based on AR, ANNs and Hybrid models.

2.1.  Introduction

2.2.  Eigenvalues and Eigenvectors

2.3.  Theorem of Singular Values Decomposition

2.4.  One-level Singular Value Decomposition of a Hankel matrix

2.4.1.         Embedding

2.4.2.         Decomposition

2.4.3.         Unembedding

2.5.  Multi-level Singular Value Decomposition of a Hankel matrix

2.5.1.         Embedding

2.5.2.         Decomposition

2.5.3.         Unembedding

2.6.  Empirical Applications

2.6.1.         Extraction of Components from traffic accidents time series based on HSVD and MSVD

2.6.2.         Extraction of Components from fishery time series based on HSVD and MSVD

3.       Forecasting based on components

3.2.  One-step ahead forecasting

3.3.  Multi-step ahead forecasting

3.3.1.         Direct Strategy

3.4.  Empirical Applications

3.4.1.         Forecasting of traffic accidents based on HSVD and MSVD

3.4.2.         Forecasting of anchovy stock based on HSVD and MSVD

3.4.3.         Forecasting of sardine stock based on HSVD and MSVD

List of Tables

List of Acronyms

References