Dunis / Middleton / Karathanasopolous Artificial Intelligence in Financial Markets

Cutting Edge Applications for Risk Management, Portfolio Optimization and Economics
1. Auflage 2016
ISBN: 978-1-137-48880-0
Verlag: Palgrave Macmillan UK
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 349 Seiten

Reihe: New Developments in Quantitative Trading and Investment

As technology advancement has increased, so to have computational applications for forecasting, modelling and trading financial markets and information, and practitioners are finding ever more complex solutions to financial challenges. Neural networking is a highly effective, trainable algorithmic approach which emulates certain aspects of human brain functions, and is used extensively in financial forecasting allowing for quick investment decision making.

This book presents the most cutting-edge artificial intelligence (AI)/neural networking applications for markets, assets and other areas of finance. Split into four sections, the book first explores time series analysis for forecasting and trading across a range of assets, including derivatives, exchange traded funds, debt and equity instruments. This section will focus on pattern recognition, market timing models, forecasting and trading of financial time series. Section II provides insights into macro and microeconomics and how AI techniques could be used to better understand and predict economic variables. Section III focuses on corporate finance and credit analysis providing an insight into corporate structures and credit, and establishing a relationship between financial statement analysis and the influence of various financial scenarios. Section IV focuses on portfolio management, exploring applications for portfolio theory, asset allocation and optimization.

This book also provides some of the latest research in the field of artificial intelligence and finance, and provides in-depth analysis and highly applicable tools and techniques for practitioners and researchers in this field. 

Dr Christian L. Dunis is a Founding Partner of Acanto Research, where he is responsible for global risk and new products. He is also Emeritus Professor of Banking and Finance at Liverpool John Moores University where he directed the Centre for International Banking, Economics and Finance (CIBEF) from February 1999 through to August 2011. Christian holds a MSc and a Superior Studies Diploma in International Economics, and a PhD in Economics from the University of Paris. Dr Peter W. Middleton completed his PhD at the University of Liverpool. His working experience is in Asset Management and he has published numerous articles on Financial Forecasting of Commodity spreads and Equity time series. Dr Andreas Karathanasopoulos studied for his MSc and Phd at Liverpool John Moores University under the supervision of Professor Christian Dunis. His working experience is academic having taught at Ulster University, London Metropolitan University and the University of East London. He is currently Associate Professor at the American University of Beirut and has published over 30 articles and one book in the area of artificial intelligence. Dr Konstantinos Theofilatos completed his MSc and PhD in the University of Patras Greece. His research interests include computational intelligence, financial time series forecasting and trading, bioinformatics, data mining and web technologies. He has published 27 publications in scientific peer reviewed journals and over 30 articles in conference proceedings.

Autoren/Hrsg.

Dunis, Christian L.

Middleton, Peter W.

Karathanasopolous, Andreas

Theofilatos, Konstantinos

Inhaltsverzeichnis

1;Preface;5
1.1;Contents;6
1.2; The Editors;10
1.3; Acknowledgements;10
1.4; Final Words;10
1.5; References;11
2;Contents;13
3;Notes on Contributors;15
4;Part I: Introduction to Artificial Intelligence;16
4.1;1: A Review of Artificially Intelligent Applications in the Financial Domain;17
4.1.1;1 Introduction;17
4.1.1.1; Applications of ANN in Finance;22
4.1.1.1.1; Portfolio Management;23
4.1.1.1.2; Stock Market Prediction;24
4.1.1.1.3; Risk Management;25
4.1.2;2 Application of Expert Systems in Finance;25
4.1.2.1; Portfolio Management;26
4.1.2.2; Stock Market Prediction;33
4.1.2.3; Risk Management;33
4.1.3;3 Applications of Hybrid Intelligence in Finance;34
4.1.3.1; Portfolio Management;34
4.1.3.2; Stock Market Prediction;37
4.1.3.3; Risk Management;38
4.1.4;4 Conclusion;42
4.1.5;5 Appendix 1;43
4.1.5.1; Regression Analysis [7];43
4.1.5.2; Classification [7];44
4.1.5.3; Clustering [7];45
4.1.5.4; Fuzzy c-means clustering [7];46
4.1.5.5; Back propagation Algorithm Code in MATLAB [111];46
4.1.5.6; Sample Code of NN Using MATLAB for Finance Management;48
4.1.5.6.1; Required functions [6];48
4.1.5.6.2; Load Historic DAX Prices;48
4.1.5.6.3; Plotting Financial Data [6];48
4.1.5.6.4; CAPM [6];49
4.1.5.6.5; Stock Price Prediction Based on Curve Fitting [6];50
4.1.6;References;51
5;Part II: Financial Forecasting and Trading;59
5.1;2: Trading the FTSE100 Index: ‘Adaptive’ Modelling and Optimization Techniques;60
5.1.1;1 Introduction;60
5.1.2;2 Literature Review;62
5.1.3;3 Related Financial Data;64
5.1.4;4 Proposed Method;66
5.1.5;5 Empirical Results;70
5.1.5.1; Benchmark Models;70
5.1.5.2; Trading Performance;71
5.1.6;6 Conclusions and Future Work;77
5.1.7;References;78
5.2;3: Modelling, Forecasting and Trading the Crack: A Sliding Window Approach to Training Neural Networks;81
5.2.1;1 Introduction;81
5.2.2;2 Literature Review;86
5.2.2.1; Modelling the Crack;86
5.2.2.2; Training of Neural Networks;87
5.2.3;3 Descriptive Statistics;88
5.2.4;4 Methodology;95
5.2.4.1; The MLP Model;95
5.2.4.2; The PSO Radial Basis Function Model;97
5.2.5;5 Empirical Results;100
5.2.5.1; Statistical Accuracy;100
5.2.5.2; Trading Performance;101
5.2.6;6 Concluding Remarks and Research Limitations;104
5.2.7;7 Appendix;107
5.2.7.1; Performance Measures;107
5.2.7.2; Supplementary Information;107
5.2.7.3; ARMA Equations and Estimations;109
5.2.7.3.1; GARCH Equations and Estimations;111
5.2.7.4; PSO Parameters;116
5.2.7.5; Best Weights over the Training Windows;116
5.2.8;References;116
5.3;4: GEPTrader: A New Standalone Tool for Constructing Trading Strategies with Gene Expression Programming;119
5.3.1;1 Introduction;119
5.3.2;2 Literature Review;120
5.3.2.1; Genetic Programming and Its Applications to Financial Forecasting;120
5.3.2.2; Gene Expression Programming and Previous Applications;121
5.3.3;3 Dataset;122
5.3.4;4 GEPTrader;123
5.3.4.1; Proposed Algorithm;123
5.3.4.2; GEPTrader Graphical User Interface;125
5.3.5;5 Empirical Results;126
5.3.5.1; Benchmark Models;126
5.3.5.2; Statistical Performance;127
5.3.5.3; Trading Performance;127
5.3.6;6 Conclusions;130
5.3.7;References;131
6;Part III: Economics;134
6.1;5: Business Intelligence for Decision Making in Economics;135
6.1.1;1 Introduction;135
6.1.2;2 Literature Review;137
6.1.2.1;General Equations for Macroeconomic Output;140
6.1.3;3 Methodology for Creating the  Business-­Automated Data Economy Model;143
6.1.4;4 Empirical Results of the Model;150
6.1.5;5 Conclusions;163
6.1.6;References;166
6.2;Part IV: Credit Risk and Analysis;169
6.3;6: An Automated Literature Analysis on Data Mining Applications to Credit Risk Assessment;170
6.3.1;1 Introduction;170
6.3.2;2 Materials and Methods;172
6.3.2.1; Search Criteria;172
6.3.2.2; Text Mining;173
6.3.2.3; Topics of Articles;175
6.3.2.4; Proposed Approach;176
6.3.3;3 Results and Analysis;178
6.3.3.1; Articles;178
6.3.3.2; Text Mining;179
6.3.3.3; Topics of Articles;181
6.3.4;4 Conclusions;183
6.3.5;References;184
6.4;7: Intelligent Credit Risk Decision Support: Architecture and Implementations;187
6.4.1;1 Introduction;187
6.4.2;2 Literature Review;188
6.4.2.1; Machine Learning Techniques;188
6.4.2.2; Techniques for Classification;190
6.4.2.3; Credit Risk Problems, Solved by Artificial Intelligence;192
6.4.3;3 Decision Support and Expert Systems for Credit Risk Domain;195
6.4.3.1; Decision Support Systems: Definitions, Goals, Premises;196
6.4.3.2; Main Types of Decision Support Systems;197
6.4.3.3; Recent Developments in Decision Support Systems for Banking Problems;200
6.4.3.4; Requirements for Credit Risk DSS;201
6.4.3.5; Financial Standards Based Decision Support;204
6.4.3.6; Developed Architecture for XBRL-Integrated DSS;207
6.4.4;4 Conclusions;211
6.4.5;References;213
6.5;8: Artificial Intelligence for Islamic Sukuk Rating Predictions;219
6.5.1;1 Introduction;219
6.5.2;2 Literature Review;221
6.5.2.1; What Is Sukuk;221
6.5.2.2; Sukuk Rating Methodology Based on Recourse of the Underlying Asset;223
6.5.2.3; Previous Studies on Rating Prediction;224
6.5.2.4; Variable Selection;226
6.5.3;3 Data and Research Method;227
6.5.3.1; Data and Sample Selection;227
6.5.3.2; Dependent and Independent Variables;227
6.5.3.3; Research Method;228
6.5.3.3.1; Multinomial Logit Regression;228
6.5.3.3.2; Decision Tree;229
6.5.3.3.3; Artificial Intelligence Neural Network;229
6.5.4;4 Result and Analysis;231
6.5.4.1; Data Screening;231
6.5.4.2; Multinomial Logistic Result;232
6.5.4.3; Decision Tree and Artificial Intelligence Neural Network Result;236
6.5.4.3.1; Phase one: General training;237
6.5.4.3.2; Phase Two: Validation test;239
6.5.4.3.3; Result comparison;241
6.5.5;5 Conclusion;243
6.5.6;6 Appendices;244
6.5.6.1;Appendix 1;244
6.5.6.2; Appendix 2;245
6.5.7;References;246
7;Part V: Portfolio Management, Analysis and Optimisation;250
7.1;9: Portfolio Selection as a Multi-period Choice Problem Under Uncertainty: An Interaction-Based Approach;251
7.1.1;1 Introduction;251
7.1.2;2 The Model;254
7.1.2.1; Agents;254
7.1.2.2; Securities and Portfolios;260
7.1.2.3; Data;261
7.1.3;3 Simulation Results;262
7.1.3.1; Baseline Framework;263
7.1.3.2; Portfolio Selection in a Bear Market;267
7.1.3.3; Portfolio Selection in a Bull Market;269
7.1.4;4 Consistency in Selection;270
7.1.4.1; Coefficient of Variation;271
7.1.4.2; Monte Carlo;273
7.1.5;5 Discussion;276
7.1.6;6 Conclusion;285
7.1.7;Appendix: Fragmented pseudo-code;286
7.1.8;References;287
7.2;10: Handling Model Risk in Portfolio Selection Using Multi-Objective Genetic Algorithm;291
7.2.1;1 Introduction;291
7.2.2;2 Portfolio Optimization and Modern Portfolio Theory;292
7.2.3;3 The Concepts of Model Risk;294
7.2.4;4 Multi-Objective Genetic Algorithms for Portfolio Optimization;296
7.2.5;5 A Portfolio’s Sharpe Ratio Error;300
7.2.6;6 Stock Forecasting Models;303
7.2.7;7 The Experiment;304
7.2.8;8 Empirical Results and Analyses;306
7.2.9;9 Conclusions;314
7.2.10;References;316
7.3;11: Linear Regression Versus Fuzzy Linear Regression: Does it Make a Difference in the Evaluation of the Performance of Mutual Fund Managers?;317
7.3.1;1 Introduction;317
7.3.2;2 Methodology;319
7.3.2.1; Treynor-Mazuy model;320
7.3.2.2; Henriksson-Merton model;321
7.3.2.3; Fuzzy Linear Regression;322
7.3.3;3 Data Set Description;325
7.3.4;4 Empirical Application;326
7.3.4.1; Results and Discussion;326
7.3.4.2; The Performance of Mutual Funds Managers;332
7.3.4.3; Fuzzy Similarity Ratios;335
7.3.5;5 Conclusions and Future Perspectives;336
7.3.6;References;338
8;Index;342