Neural Network Time Series: Forecasting of Financial Markets [Hardcover]

E. Michael Azoff (Author)

Book Description

ISBN-10: 0471943568 | ISBN-13: 978-0471943563 | Publication Date: September 13, 1994 | Edition: 1

Neural Network Time Series Forecasting of Financial Markets E. Michael Azoff The first comprehensive and practical introduction to using neural networks in financial time series forecasting. This practical working guide shows you how to understand, design and profitably use neural network techniques in financial market forecasting. It encompasses:

• A tutorial introduction to neural networks
• Data preprocessing
• Key network design issues
• Random walk probability theory
• Fully specified benchmarks (and code for implementing the benchmarks as pre-trained networks)
• An overview of futures trading
• Discussion of trading systems and risk management

The book focuses on the multilayer perception, one of the most powerful and successful network architectures that is used in the majority of commercial applications, especially financial time series forecasting. The fully specified benchmarks are a unique feature of the book and will be of particular benefit if you are contemplating designing your own neural network using one of the many commercial simulators.

Editorial Reviews

From the Publisher

Introduces the use of neural networks in forecasting and, in particular, financial time series forecasting. Provides much-needed guidance for applying predictive and decision-enhancing functions of neural nets to a wide range of global capital markets investments and futures trading.

From the Inside Flap

Neural Network Time Series Forecasting of Financial Markets A neural network is a computer program that can recognise patterns in data, learn from this and (in the case of time series data) make forecasts of future patterns. There are now over 20 commercially available neural network programs designed for use on financial markets and there have been some notable reports of their successful application. However, like any other computer program, neural networks are only as good as the data they are given and the questions that are asked of them. Proper use of a neural network involves spending time understanding and cleaning the data: removing errors, preprocessing and postprocessing. This book takes the reader beyond the ‘black-box’ approach to neural networks and provides the knowledge that is required for their proper design and use in financial markets forecasting —with an emphasis on futures trading. Comprehensively specified benchmarks are provided (including weight values), drawn from time series examples in chaos theory and financial futures. The book covers data preprocessing, random walk theory, trading systems and risk analysis. It also provides a literature review, a tutorial on backpropagation, and a chapter on further reading and software. For the professional financial forecaster this book is without parallel as a comprehensive, practical and up-to-date guide to this important subject.

See all Editorial Reviews

Product Details

• Hardcover: 212 pages
• Publisher: John Wiley & Sons; 1 edition (September 13, 1994)
• Language: English
• ISBN-10: 0471943568
• ISBN-13: 978-0471943563
• Product Dimensions: 9.2 x 6 x 0.7 inches
• Shipping Weight: 1 pounds
• Average Customer Review: 5.0 out of 5 stars  See all reviews (1 customer review)
• Amazon Best Sellers Rank: #1,440,870 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

19 of 20 people found the following review helpful:

5.0 out of 5 stars Better than I expected, February 25, 1997

By A Customer

This review is from: Neural Network Time Series: Forecasting of Financial Markets (Hardcover)

A good introduction to neural nets and their applicability tothe futures markets. Provides mathematical/theoretical basis fortechniques involved in neural net training, testing, chaos and touches on nonlinear systems in general. Also provides interesting benchmarks of several nets against several time series. Great bibliography. Includes Fortran source code for running trained nets but not for training a net.