Ordinary shares, exotic methods : financial forecasting using data mining techniques / Francis Eng-Hock Tay, Lixiang Shen, Lijuan Cao.

Format:

Book

Author/Creator:

Tay, Francis E. H.

Contributor:

Cao, Lijuan.

Shen, Lixiang.

Language:

English

Subjects (All):

Data mining.

Investments--Data processing.

Investments.

Stock price forecasting--Data processing.

Stock price forecasting.

Physical Description:

1 online resource (198 p.)

Place of Publication:

River Edge, N.J. : World Scientific, c2003.

Language Note:

English

Summary:

Exotic methods refer to specific functions within general soft computing methods such as genetic algorithms, neural networks and rough sets theory. They are applied to ordinary shares for a variety of financial purposes, such as portfolio selection and optimization, classification of market states, forecasting of market states and data mining. This is in contrast to the wide spectrum of work done on exotic financial instruments, wherein advanced mathematics is used to construct financial instruments for hedging risks and for investment.In this book, particular aspects of the general method are

Contents:

Contents; 1 Financial Forecasting Problem and Data Mining Techniques; 1.1. Introduction; 1.2. Typical Applications in Data Mining; 1.3. Financial Forecasting Problem; 1.4. Genetic Niching, Rough Sets and Support Vector Machines; 2 Genetic Algorithms and Genetic Niching; 2.1. Simple Genetic Algorithms; 2.2. Niching Methods for Multimodal Optimization; 2.2.1. Sharing method; 2.2.2. Crowding method; 2.2.3. Clearing method; 2.2.4. Other niching methods; 2.3. Performance Criteria for Multimodal Optimization; 2.3.1. Maximum peak ratio; 2.3.2. Chi-square like performance statistic

2.4. Multimodal Functions for Test2.5. Genetic Procedures for Multimodal Optimization; 2.6. Results; 2.6.1. Test on F1 function; 2.6.2. Test on F2 function; 2.6.3. Test on F3 function; 2.7. Summary; 3 Portfolio Selection and Optimization Using Genetic Operators; 3.1. Introduction; 3.2. Literature Review; 3.3. Clearing Method for Genetic Niching; 3.4. Genetic Operators for Portfolio Optimization; 3.5. Generating Multiple Portfolios; 3.6. Transaction Costs; 3.7. Portfolio Optimization with Transaction Costs; 3.8. Conclusion

4 The Rough Sets Theory Basics and Its Applications in Economic and Financial Forecasting4.1. Introduction; 4.2. Rough Sets Basics; 4.2.1. Information system and decision table; 4.2.2. Lower and upper approximation; 4.2.3. Quality of approximation; 4.2.4. The discernibility matrices and discernibility function; 4.2.5. Reduct and core of attributes; 4.2.6. Decision rules; 4.3. Financial Applications using the Rough Sets Theory; 4.3.1. Business failure prediction; 4.3.2. Database marketing; 4.3.3. Financial investment; 5 Time Series Forecasting using Rough Sets Theory; 5.1. Introduction

5.2. Temporal Rule Discovery Problem5.3. Temporal Information System (TIS); 5.4. Converting Time Series to Rough Sets Objects; 5.4.1. The mobile window; 5.4.2. ""Columnizing""; 5.5. Financial Market Prediction; 5.6. The Experiment Process; 5.6.1. Indicators study; 5.6.2. RoughSOM system; 5.7. Data Preparation; 5.8. Rules Extraction; 5.9. Results of Discussion; 5.10. Summary; 6 A Review of Support Vector Machines in Regression Estimation; 6.1. Introduction; 6.2. Theory of SVMs in Regression Estimation; 6.2.1. ε-lnsensitive Loss Function; 6.2.2. Linear SVMs; 6.2.3. Nonlinear SVMs

6.3. Training Algorithms of SVMs6.4. Methodologies; 6.5. Applications and Performance; 6.6. Conclusions and Future Works; 7 Application of Support Vector Machines in Financial Time Series Forecasting; 7.1. Introduction; 7.2. Literature Review in Financial Forecasting; 7.3. Data Sets and Data Preprocessing; 7.4. Prediction Performance Criteria; 7.5. Experimental Results; 7.5.1. Prediction Results; 7.5.2. Trading Simulation and Results; 7.6. Conclusions; 8 Other Methods and Their Applications; 8.1. Expert System; Rules, Predicates, Semantic Nets, Frames, Objects.; 8.2. Fuzzy Logic

Investor Classification

Notes:

Description based upon print version of record.

Includes bibliographical references (p. 155-183) and index.

ISBN:

9786611933890

9781281933898

1281933899

9789812791375

981279137X