لایب سان

Contents List of figures ix List of tables xi List of contributors xiii Preface xv Acknowledgements xix PART I Introduction 1 1 Computational intelligence: recent advances, perspectives and open problems 3 K O N S T A N T I N O S T H E O F I L A T O S , E F S T R A T I O S G E O R G O P O U L O S , S P I R O S L I K O T H A N A S S I S A N D S E F E R I N A M A V R O U D I 2 Financial forecasting and trading strategies: a survey 22 C H A R A L A M P O S S T A S I N A K I S A N D G E O R G I O S S E R M P I N I S PART II Trading and investments with traditional computational intelligence techniques 41 3 Hidden Markov models: financial modelling and applications 43 S O V A N M I T R A 4 Adaptive filtering on forecasting financial derivatives indices 66 C H R I S T O S D I M I T R A K O P O U L O S , A N D R E A S K A R A T H A N A S O P O U L O S , G E O R G I O S S E R M P I N I S A N D S P I R O S L I K O T H A N A S S I S viii Contents PART III Trading and investments with artificial neural networks 79 5 Modelling and trading the corn–ethanol crush spread with neural networks 81 C H R I S T I A N L . D U N I S , J A S O N L A W S , P E T E R W . M I D D L E T O N A N D A N D R E A S K A R A T H A N A S O P O U L O S 6 Trading decision support with historically consistent neural networks 116 H A N S - J ö R G V O N M E T T E N H E I M PART IV Trading and investments with hybrid evolutionary methodologies 131 7 Advanced short-term forecasting and trading deploying neural networks optimized with an adaptive evolutionary algorithm 133 K O N S T A N T I N O S T H E O F I L A T O S , T H O M A S A M O R G I A N I O T I S , A N D R E A S K A R A T H A N A S O P O U L O S , G E O R G I O S S E R M P I N I S , E F S T R A T I O S G E O R G O P O U L O S A N D S P I R O S L I K O T H A N A S S I S 8 Using argumentation and hybrid evolutionary multi- model partitioning algorithms for efficient portfolio construction 146 N I K O L A O S S P A N O U D A K I S , K O N S T A N T I N A P E N D A R A K I A N D G R I G O R I O S B E L I G I A N N I S PART V Trading and investments with advanced computational intelligence modelling techniques 175 9 Forecasting DAX 30 using support vector machines and VDAX 177 R A F A E L R O S I L L O , J A V I E R G I N E R A N D D A V I D D E L A F U E N T E 10 Ensemble learning of high-dimensional stock market data 192 M A N O L I S M A R A G O U D A K I S A N D D I M I T R I O S S E R P A N O S Index 216 Figures 1.1 The MLP neural network architecture 6 1.2 The RNN architecture 6 1.3 The HONN architecture 8 1.4 The RBF neural network architecture 9 1.5 Meta- heuristic algorithms flow chart 13 3.1 A diagram of a two-state Markov model with state transitions depicted by arrows 44 3.2 Graphs for the S&P 500 index 1976–1996 57 3.3 Graphs for the Nikkei 225 index 1976–1996 58 3.4 Graph of Bank of England interest rates (%) for each month from October 1980 to December 2008 59 3.5 Graph of interest rates states for each month from October 1980 to December 2008 60 5.1 The corn–ethanol crush CBOT daily closing prices 89 5.2 Histogram of corn–ethanol spread return series 90 5.3 A single output, inter-connected MLP model 96 5.4 Elman RNN architecture with two neurons/nodes for the hidden layer 97 5.5 Second order HONN with three inputs 99 6.1 Modelling with the HCNN. The initial hidden states can also be trained 118 8.1 The structure of the hybrid evolutionary multi-model partitioning algorithm used for financial time series forecasting 158 8.2 Mapping from a fixed dimensional internal representation to a variable length NARMAX parameter vector. The resulting order is n(p, q, r) = (3, 2, 3) 159 8.3 The fitness of each candidate model is the model conditional pdf 160 8.4 The PORTRAIT system architecture 163 8.5 The investment process using the PORTRAIT tool 164 8.6 A screenshot of the PORTRAIT tool 165 8.7 The bear market context, the moderate investor role and the specific context that combines them returns for all eight years 167 x Figures 8.8 The compound return of all scenarios sorted according to the compound return of 2008, compared to the compound return of the RASE and the average return of all MFs 168 9.1 An example of how a kernel function works 178 9.2 DAX (left axis) and VDAX (right axis) 180 9.3 Design of the trading rule 182 9.4 Cross correlation DAX returns and VDAX changes 184 9.5 Cumulative DAX 30 points for each strategy from 2000 to 2011 188 10.1 Hierarchical decomposition of a random forests classifier on a dataset 199 10.2 A simple Bayesian network depicting the interelation of five random variables 202 10.3 An example of a Bayesian network with the Markov blanket of node x 202 10.4 Plot of virtual portfolio balance of two different trading strategies 212

برای ارسال نظر لطفا وارد شوید یا ثبت نام کنید