There're 3 approaches for stock market prediction -- technical analysis (TA), fundamental analysis (FA) and efficient market hypothesis (EMH).  TA uses historical price data (open, high, low, close, volume) while FA uses company financial statements.  EMH rules out TA and FA by stating that stock market is so efficient that all information is contained in current stock price.  One can disprove EMH if one can make excess returns through TA or FA.

With rapid computer hardware progress, one can test and build trading system easily now.  Basically speaking, there're two categories of trading models -- parametric and non-parametric models.  Parametric model includes traditional TA indices (RSI, MACD, KDJ...) based rules and statistical models, such as ARMA (Autoregressive Moving Average), GARCH (General Autoregressive Conditional Heteroscedasticity).  Non-parametric model usually relates to artificial intelligent technology, such as artificial neural network, genetic algorithm and nearest-neighbor instance-based learning.  I'm particularly interested in feedforward backpropagation neural network and think it's a key technique to conquer complex market behaviors.

Some researchers suggest that stock market is non-linear chaotic system.  Currently, although we can't use chaos theory to predict stock market directly, we can use chaotic analysis to get some hints.  Lyapunov exponents tell us if a system is chaotic or not.  Correlation dimension gives the least dimension of one lagged dependent time series variable to represent all dependent and independent variables for a chaotic system.  Hurst exponent shows predictability of time series data.

Papers

Neural network for financial market and time series

1. A study on training criteria for financial time series forecasting
2. A Comparison Between Neural-Network Forecasting Techniques-Case Study: River Flow Forecasting
3. Applying Artificial Neural Networks to Business, Economics and Finance
4. A commodity trading model based on neural network-expert system hybrid
5. Embedding Technical Analysis into Neural Network Based Trading Systems
6. Noisy Time Series Prediction using a Recurrent Neural Network and grammatical inference
7. Mining Sales Data using a Neural Network Model of Market Response
8. Guidelines for Financial Forecasting with Neural Networks
9. A comparative study on feedforward and recurrent neural networks in time series prediction using gradient descent learning
10. Improving the Accuracy of Financial Time Series Prediction Using Ensemble Networks and High Order Statistics
11. Prediction Of Stock Market Index Changes
12. Stock Price Prediction using Neural Networks
13. Feedforward and Recurrent Neural Networks and Genetic Programs for stock market and time series forecasting
14. Neural Networks, Financial Trading and the Efficient Markets
15. Neural Networks for Time Series Processing
16. Selection of Learning Algorithms for Trading Systems Based on Biased Estimators
17. Parallel Back-Propagation for the Prediction of Time Series
18. Improving The Prediction Accuracy Of Financial Time Series By Using Multi-Neural Network Systems and Enhanced Data Preprocessing
19. The Use of Parsimonious Neural Networks for forecasting financial time series
20. Time Series Prediction and Neural Networks
21. Using Neural Networks And Genetic Algorithms to Predict Stock Market Returns
22. An Empirical Analysis of Data Requirements for Financial Forecasting with Neural Networks
23. A case study on using neural networks to perform technical forecasting of forex
24. Equity Forecasting: A Case Study On The KLSE Index
25. Foreign Exchange Rates Forecasting with Neural Networks
26. Neural Networks For Technical Analysis: A study on KLCI
27. On Developing a Financial Prediction System: Pitfalls and Possibilities
28. Integrating Ensemble of Intelligent Systems for Modeling Stock Indices

Neural network benchmark

1. PROBEN1 - A Set of Neural Network Benchmark Problems and Benchmarking Rules
2. Neural Network vs. ARMA Modelling: constructing benchmark case studies of river flow prediction

Neural network design

1. Confidence and prediction intervals for neural network ensembles
2. Searching for optimal MLP
3. Global Optimization for Artificial Neural Networks: A Tabu Search Application
4. Input Window Size and Neural Network Predictors
5. Performance and Efficiency: Recent Advances in Supervised Learning
6. Constructive Feedforward Neural Networks for Regression Problems
7. Evaluating Neural Network Predictors by Bootstrapping
8. Neural Network Implementation in SAS Software
9. One step ahead forecasting using Multilayered perceptron
10. Optimal learning in artificial neural networks: a theoretical view
11. A comparison of some error estimates for neural network models
12. Handling Time-Warped Sequences with Neural Networks
13. Heuristic Principles For The Design Of Artificial Neural Networks
14. Understanding Neural Networks as Statistical Tools
15. Ensembling Neural Networks: Many Could Be Better Than All

Chaos

1. TIME SERIES PREDICTION USING SUPERVISED LEARNING AND TOOLS FROM CHAOS THEORY
2. Modeling Chaotic Behavior of Stock Indices Using Intelligent Paradigms
3. Distribution of Mutual Information
4. Dynamic Modeling of Chaotic Time Series with Neural Networks
5. Prediction of Chaotic Time Series with Neural Networks
6. A Hybrid Financial Trading System Incorporating Chaos Theory, Statistical and Artificial Intelligence/Soft Computing methods
7. Time Series Prediction and Neural Networks
8. EPNet for Chaotic Time-Series Prediction

Genetic Algorithm for neural network design

1. An Indexed Bibliography of Genetic Algorithms and Neural Networks
2. Evolutionary Design of Neural Architectures - A Preliminary Taxonomy and Guide to Literature
3. Evolving Networks: Using the Genetic Algorithm with Connectionist Learning
4. Evolutionary Algorithms for Neural Network Design and Training
5. Efficient Evolution of Neural Network Topologies
6. Efficient Reinforcement Learning through Evolving Neural Network Topologies
7. Hierarchical Evolution of Neural Networks
8. Is Evolutionary Design the Solution for Optimizing Neural Networks?
9. Forming Neural Networks through Efficient and Adaptive Coevolution
10. Neuro-Evolution Through Augmenting Topologies Applied To Evolving Neural Networks to Play Othello
11. Evolutionary Design Of Neural Networks: Application To River Flow Prediction
12. Toward Global Optimization Of Neural Networks: A Comparison Of the Genetic Algorithm and Backprogpagation
13. Optimization of Neural Networks: A Comparative Analysis of the Genetic Algorithm and Simulated Annealing
14. Evolving Neural Networks through Augmenting Topologies
15. Towards Designing Neural Network Ensembles by Evolution
16. Would Evolutionary Computation Help in Designs of Artificial Neural Nets in Forecasting Financial Time Series?
17. Ensemble of GA based Selective Neural Network Ensembles
18. Evolutionary Artificial Neural Networks
19. A Review of Evolutionary Artificial Neural Networks
20. A New Evolutionary System for Evolving Artificial Neural Networks
21. Evolving Artificial Neural Networks
22. Combining Regression Estimators: GA-Based Selective Neural Network Ensemble
23. Genetic Algorithm based Selective Neural Network Ensemble
24. Evolving Fault-Tolerant Neural Networks

GA for finance

1. An Indexed Bibliography of Genetic Algorithms in Economics
2. Modeling Speculators with Genetic Programming
3. Trading Restrictions, Speculative Trades and Price Volatility: An Application of Genetic Programming
4. The Main Ingredients of Simple Trading Models for Use in Genetic Algorithm Optimization
5. Option Pricing with Genetic Programming
6. Genetic Algorithms with collective sharing for Robust Optimization in Financial Applications
7. Pricing Financial Derivatives with Genetic Programming

Finance and Time series

1. Detecting Concept Drift in Financial Time Series Prediction using Symbolic Machine Learning
2. Inferring Information from Trading
3. Stock Market Prices Do Not Follow Random Walks: Evidence from a Simple Specification Test
4. Predictable Patterns in Stock Returns
5. Predicting the Stock Market
6. Financial Returns and Efficiency as seen by an Artificial Technical Analyst
7. The Informational Role of Stock and Option Volume

News for financial market prediction

1. Text Processing for Classification
2. Currency Exchange Rate Forecasting from News Headlines
3. Using News Articles to Predict Stock
4. Daily Stock Market Forecast from Textual Web Data
5. Knowledge Discovery From Distributed And Textual Data - Cho
6. Daily Prediction of Major Stock Indices from textual.. - Wüthrich
7. Learned Text Categorization By Backpropagation Neural Network - Yin, SAVIO
8. Discovering Probabilistic Decision Rules - Wüthrich

Books

AI for financial market

1. Neural networks in finance and investing, HG4012.5.N48 1993
2. Intelligent systems and financial forecasting, HG4515.5.K56 1997
3. Neural network time series forecasting of financial markets, HG6024.A3 A96 1994
4. Neural networks for financial forecasting, HG4515.5.G37 1996
5. Expert trading systems with kernel regression, HG4523.w65 2000
6. Learning kernel classifiers, Q325.5.H48 2002
7. Intelligent systems for finance and business, HD30.2.I554 1995
8. Visual explorations in finance, HG4012.5.v57 1998
9. Neural networks in the capital markets, HG4523.P47 1991

Market Trading

1. Trading systems and methods, HG6046.K34 1998
2. Market models, HG4515.3.A38 2001
3. Trading on the edge, HG4012.5.T7 1994

Chaos

1. Analysis of observed chaotic data, Q172.5.C45 A23 1996
2. Chaos and order in the capital markets : a new view of cycles, prices, and market volatility, HG4523.P47 1991
3. Dynamics: Numerical Explorations, QA1.A647 v.101
4. Trading chaos, HG4523.w554 1995

Neural Network

1. Masters, a c++ source book for neural network, QA76.87.M367 1995
2. Principal component neural network, QA76.87.D53 1996
3. Industrial applications of neural networks, QA76.87.I523 1999

MATLAB

1. MATLAB: an introduction with applications, QA297.G48 2004
2. Numerical methods in finance: a MATLAB-Based introduction, HG176.5.B73 2002
3. MATLAB guide, QA297.H5217 2000
4. Computational Statistics Handbook with MATLAB, QA276.4.M272 2001

Hurst exponent and neural network prediction

Paper, matlab codes

Stock market prediction with multiple classifiers

Paper, matlab codes