Advances in Time Series Forecasting: Volume 2

This volume is a valuable source of recent knowledge about advanced time series forecasting techniques such as artificial neural networks, fuzzy time series, or hybrid approaches. New forecasting frameworks are discussed and their application is demonstrated. The second volume of the series includes applications of some powerful forecasting approaches with a focus on fuzzy time series methods. Chapters integrate these methods with concepts such as neural networks, high order multivariate systems, deterministic trends, distance measurement and much more. The chapters are contributed by eminent scholars and serve to motivate and accelerate future progress while introducing new branches of time series forecasting. This book is a valuable resource for MSc and PhD students, academic personnel and researchers seeking updated and critically important information on the concepts of advanced time series forecasting and its applications. CONTENTS 6 PREFACE 9 List of Contributors 11 Fuzzy Time Series Forecasting Models Evaluation Based on A Novel Distance Measure 12 Cagdas Hakan Aladag1,* and I. Burhan Turksen2 12 INTRODUCTION 12 THE PROPOSED DISTANCE MEASURE AND THE SUGGESTED PERFORMANCE CRITERION 16 THE APPLICATION 21 CONCLUDING REMARKS 30 CONFLICT OF INTEREST 32 ACKNOWLEDGEMENTS 32 REFERENCES 32 A New Fuzzy Time Series Forecasting Model with Neural Network Structure 35 Eren Bas* and Erol Egrioglu 35 INTRODUCTION 35 PROPOSED METHOD 37 APPLICATION 41 CONCLUSIONS AND DISCUSSIONS 43 CONFLICT OF INTEREST 43 ACKNOWLEDGEMENTS 44 REFERENCES 44 Two Factors High Order Non Singleton Type-1 and Interval Type-2 Fuzzy Systems for Forecasting Time Series with Genetic Algorithm 48 M.H. Fazel Zarandi1, *, M. Yalinezhaad1 and I.B. Turksen2 48 INTRODUCTION 48 Interval Type-2 Fuzzy Logic Sets and Systems 51 Type-2 Fuzzy Logic Sets 51 Non Singleton Interval Type-2 Fuzzy Logic Systems 53 Determination of Footprints of Uncertainty (Umf and Lmf) in Interval Type-2 Fuzzy Logic Sets 54 Fundamental Concepts of Fuzzy Time Series 56 Proposed Two Factors High Order Non Singletontype-1 and Interval Type-2 Fuzzy Time Series Systems 56 Tuning Method for Type-1 and Interval Type-2 FTSs with Genetic Algorithm 59 Experimental Results by Temperature Prediction and TAIEX Forecasting 61 Temperature Prediction with Proposed Method 61 TAIEX Forecasting By Applying the Proposed Method with Genetic Algorithm 72 GA Procedure 79 Selection and Pairing 80 Crossover 80 Mutation and Reinsertion 80 Termination Condition 80 Type Reduction and Defuzzification 80 CONCLUSION AND FUTURE WORKS 84 CONFLICT OF INTEREST 84 ACKNOWLEDGEMENTS 84 REFERENCES 84 A New Neural Network Model with Deterministic Trend and Seasonality Components for Time Series Forecasting 87 Erol Egrioglu1,*, Cagdas Hakan Aladag2, Ufuk Yolcu3, Eren Bas1 and Ali Z. Dalar1 87 INTRODUCTION 87 CLASSICAL TIME SERIES FORECASTING MODELS 88 ARTIFICIAL NEURAL NETWORKS FOR FORECASTING TIME SERIES 91 A NEW ARTIFICIAL NEURAL NETWORK WITH DETERMINISTIC COMPONENTS 96 APPLICATIONS 100 CONCLUSION 102 CONFLICT OF INTEREST 102 ACKNOWLEDGEMENTS 103 REFERENCES 103 A Fuzzy Time Series Approach Based on Genetic Algorithm with Single Analysis Process 104 Ozge Cagcag Yolcu* 104 INTRODUCTION 104 FUZZY TIME SERIES 107 RELATED METHODS 108 Genetic Algorithm (GA) 108 Single Multiplicative Neuron Model 109 PROPOSED METHOD 111 APPLICATIONS 115 CONCLUSION AND DISCUSSION 118 CONFLICT OF INTEREST 118 ACKNOWLEDGEMENTS 118 REFERENCES 118 Forecasting Stock Exchanges with Fuzzy Time Series Approach Based on Markov Chain Transition Matrix 122 Cagdas Hakan Aladag1,* and Hilal Guney2 122 INTRODUCTION 122 FUZZY TIME SERIES 124 TSAUR ‘S FUZZY TIME SERIES MARKOV CHAIN MODEL 125 THE IMPLEMENTATION 128 CONCLUSION 135 CONFLICT OF INTEREST 135 ACKNOWLEDGEMENTS 135 REFERENCES 135 A New High Order Multivariate Fuzzy Time Series Forecasting Model 138 Ufuk Yolcu* 138 INTRODUCTION 138 RELATED METHODOLOGY 140 The Fuzzy C-Means (FCM) Clustering Method 140 Single Multiplicative Neuron Model Artificial Neural Network (SMN-ANN) 141 Fuzzy Time Series 142 THE PROPOSED METHOD 144 APPLICATIONS 147 CONCLUSIONS AND DISCUSSION 151 CONFLICT OF INTEREST 151 ACKNOWLEDGEMENTS 152 REFERENCES 152 Fuzzy Functions Approach for Time Series Forecasting 155 Ali Z. Dalar1,*, Erol Egrioglu1, Ufuk Yolcu2 and Cagdas Hakan Aladag3 155 INTRODUCTION 155 TYPE-1 FUZZY FUNCTIONS APPROACH 157 IMPLEMENTATION 159 Australian Beer Consumption Time Series 159 Turkey Electricity Consumption Time Series 162 CONCLUSIONS 164 CONFLICT OF INTEREST 164 ACKNOWLEDGEMENTS 164 REFERENCES 164 Recurrent ANFIS for Time Series Forecasting 167 Busenur Sarıca1,*, Erol Eğrioğlu2 and Barış Aşıkgil3 167 INTRODUCTION 167 RECURRENT ADAPTIVE NETWORK FUZZY INFERENCE SYSTEMS 168 APPLICATION 171 CONCLUSION 174 CONFLICT OF INTEREST 174 ACKNOWLEDGEMENTS 174 REFERENCES 174 A Hybrid Method for Forecasting of Fuzzy Time Series 176 Eren Bas* 176 INTRODUCTION 176 THE METHODS USED IN THIS STUDY 177 Fuzzy Time Series 177 Genetic Algorithm 178 Differential Evolution Algorithm 178 PROPOSED METHOD 179 APPLICATION 182 Analysis of Canadian Lynx Data 183 CONCLUSIONS 184 CONFLICT OF INTEREST 184 ACKNOWLEDGEMENTS 184 REFERENCES 184 SUBJECT INDEX 188