Konar / Bhattacharya Time-Series Prediction and Applications

A Machine Intelligence Approach
1. Auflage 2017
ISBN: 978-3-319-54597-4
Verlag: Springer Nature Switzerland
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, Band 127, 242 Seiten

Reihe: Intelligent Systems Reference Library

This book presents machine learning and type-2 fuzzy sets for the prediction of time-series with a particular focus on business forecasting applications. It also proposes new uncertainty management techniques in an economic time-series using type-2 fuzzy sets for prediction of the time-series at a given time point from its preceding value in fluctuating business environments. It employs machine learning to determine repetitively occurring similar structural patterns in the time-series and uses stochastic automaton to predict the most probabilistic structure at a given partition of the time-series. Such predictions help in determining probabilistic moves in a stock index time-series

Primarily written for graduate students and researchers in computer science, the book is equally useful for researchers/professionals in business intelligence and stock index prediction. A background of undergraduate level mathematics is presumed, although not mandatory, for most of the sections. Exercises with tips are provided at the end of each chapter to the readers’ ability and understanding of the topics covered.

Autoren/Hrsg.

Konar, Amit

Bhattacharya, Diptendu

Inhaltsverzeichnis

1;Preface;7
2;Acknowledgements;11
3;Contents;12
4;About the Authors;15
5;1 An Introduction to Time-Series Prediction;17
5.1;Abstract;17
5.2;1.1 Defining Time-Series;17
5.3;1.2 Importance of Time-Series Prediction;18
5.4;1.3 Hindrances in Economic Time-Series Prediction;19
5.5;1.4 Machine Learning Approach to Time-Series Prediction;20
5.6;1.5 Scope of Machine Learning in Time-Series Prediction;22
5.7;1.6 Sources of Uncertainty in a Time-Series;27
5.8;1.7 Scope of Uncertainty Management by Fuzzy Sets;28
5.9;1.8 Fuzzy Time-Series;31
5.9.1;1.8.1 Partitioning of Fuzzy Time-Series;33
5.9.2;1.8.2 Fuzzification of a Time-Series;35
5.10;1.9 Time-Series Prediction Using Fuzzy Reasoning;38
5.11;1.10 Single and Multi-Factored Time-Series Prediction;42
5.12;1.11 Scope of the Book;44
5.13;1.12 Summary;45
5.14;References;48
6;2 Self-adaptive Interval Type-2 Fuzzy Set Induced Stock Index Prediction;54
6.1;Abstract;54
6.2;2.1 Introduction;55
6.3;2.2 Preliminaries;59
6.4;2.3 Proposed Approach;60
6.4.1;2.3.1 Training Phase;62
6.4.1.1;2.3.1.1 Partitioning of Main Factor Close Prices into p Intervals of Equal Length;64
6.4.1.2;2.3.1.2 Construction of IT2 or Type-1 Fuzzy Sets as Appropriate for Each Interval of Close Price;65
6.4.1.3;2.3.1.3 Fuzzy Prediction Rule (FPR) Construction for Consecutive {\varvec c(t) } s;67
6.4.1.4;2.3.1.4 Grouping of IT2/T1 Fuzzy Implications for Individual Main Factor Variation {\varvec V_{M}^{d} } (t);68
6.4.1.5;2.3.1.5 Computing Composite Secondary Variation Series (CSVS) and Its Partitioning;69
6.4.1.6;2.3.1.6 Determining Secondary to Main Factor Variation Mapping;70
6.4.2;2.3.2 Prediction Phase;71
6.4.3;2.3.3 Prediction with Self-adaptive IT2/T1 MFs;74
6.5;2.4 Experiments;75
6.5.1;2.4.1 Experimental Platform;76
6.5.2;2.4.2 Experimental Modality and Results;76
6.5.2.1;2.4.2.1 Policies Adopted;76
6.5.2.2;2.4.2.2 MF Selection;77
6.5.2.3;2.4.2.3 Adaptation Cycle;77
6.5.2.4;2.4.2.4 Varying d;80
6.6;2.5 Performance Analysis;80
6.7;2.6 Conclusion;82
6.8;2.7 Exercises;83
6.9;Appendix 2.1;90
6.10;Appendix 2.2: Source Codes of the Programs;97
6.11;References;115
7;3 Handling Main and Secondary Factors in the Antecedent for Type-2 Fuzzy Stock Prediction;119
7.1;Abstract;119
7.2;3.1 Introduction;119
7.3;3.2 Preliminaries;121
7.4;3.3 Proposed Approach;123
7.4.1;3.3.1 Method-I: Prediction Using Classical IT2FS;124
7.4.2;3.3.2 Method-II: Secondary Factor Induced IT2 Approach;126
7.4.3;3.3.3 Method-III: Prediction in Absence of Sufficient Data Points;128
7.4.4;3.3.4 Method-IV: Adaptation of Membership Function in Method III to Handle Dynamic Behaviour of Time-Series [47–52];134
7.5;3.4 Experiments;135
7.5.1;3.4.1 Experimental Platform;135
7.5.2;3.4.2 Experimental Modality and Results;136
7.6;3.5 Conclusion;141
7.7;Appendix 3.1: Differential Evolution Algorithm [36, 48–50];141
7.8;References;143
8;4 Learning Structures in an Economic Time-Series for Forecasting Applications;147
8.1;Abstract;147
8.2;4.1 Introduction;147
8.3;4.2 Related Work;150
8.4;4.3 DBSCAN Clustering—An Overview;151
8.5;4.4 Slope-Sensitive Natural Segmentation;153
8.5.1;4.4.1 Definitions;154
8.5.2;4.4.2 The SSNS Algorithm;157
8.6;4.5 Multi-level Clustering of Segmented Time-Blocks;159
8.6.1;4.5.1 Pre-processing of Temporal Segments;159
8.6.2;4.5.2 Principles of Multi-level DBSCAN Clustering;160
8.6.3;4.5.3 The Multi-level DBSCAN Clustering Algorithm;162
8.7;4.6 Knowledge Representation Using Dynamic Stochastic Automaton;163
8.7.1;4.6.1 Construction of Dynamic Stochastic Automaton (DSA);166
8.7.2;4.6.2 Forecasting Using the Dynamic Stochastic Automaton;168
8.8;4.7 Computational Complexity;170
8.9;4.8 Prediction Experiments and Results;172
8.10;4.9 Performance Analysis;173
8.11;4.10 Conclusion;176
8.12;Appendix 4.1: Source Codes of the Programs;177
8.13;References;201
9;5 Grouping of First-Order Transition Rules for Time-Series Prediction by Fuzzy-Induced Neural Regression;203
9.1;Abstract;203
9.2;5.1 Introduction;203
9.3;5.2 Preliminaries;207
9.3.1;5.2.1 Fuzzy Sets and Time-Series Partitioning;207
9.3.2;5.2.2 Back-Propagation Algorithm;208
9.3.3;5.2.3 Radial Basis Function (RBF) Networks;209
9.4;5.3 First-Order Transition Rule Based NN Model;210
9.5;5.4 Fuzzy Rule Based NN Model;215
9.6;5.5 Experiments and Results;218
9.6.1;5.5.1 Experiment 1: Sunspot Time-Series Prediction;218
9.6.2;5.5.2 Experiment 2: TAIEX Close-Price Prediction;223
9.7;5.6 Conclusion;227
9.8;Appendix 5.1: Source Codes of the Programs;230
9.9;References;246
10;6 Conclusions;248
10.1;6.1 Conclusions;248
10.2;6.2 Future Research Directions;249
11;Index;250