Financial Prediction Using Neural Networks [Hardcover]

Joseph S. Zirilli (Author)

Book Description

ISBN-10: 1850322341 | ISBN-13: 978-1850322344 | Publication Date: December 1996

Research into the use of neural nets for financial prediction is a major research interest at present. This book will focus on approaches to performing trend analysis through the use of neural nets, comparing the results of experiments on various types of markets, and will include a review of current work in the area. It should appeal to students in both neural computing and finance as well as to financial analysts and academic and professional researchers in the field of neural network applications.

Product Details

• Hardcover: 135 pages
• Publisher: Intl Thomson Computer Pr (Sd) (December 1996)
• Language: English
• ISBN-10: 1850322341
• ISBN-13: 978-1850322344
• Product Dimensions: 9.3 x 6.3 x 0.6 inches
• Shipping Weight: 13.6 ounces
• Average Customer Review: 3.0 out of 5 stars  See all reviews (2 customer reviews)
• Amazon Best Sellers Rank: #2,918,533 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

3 of 5 people found the following review helpful:

2.0 out of 5 stars Joe Zirilli knows how to make money...., August 14, 2005

By 

Reza Mostafid - See all my reviews
(REAL NAME)   

This review is from: Financial Prediction Using Neural Networks (Hardcover)

....simply by writing a little monoghraph and selling it for 160 bucks....I am not an academic snub, but rather a hardworking engineer that works for a financial firm helping to harness neural networks and traditional forecasting methods for the preditction of stock markets in the middle east.

I give Mr. Zirilli's book 2 stars, because he does share some insights about how to go on about predicting trends and future stock market prices.

HOWEVER having said that, I ordered the book as an electronic copy ( NO I am not mistaken..same ISBN same no of pages etc ) for only 33 bucks and as I said it served me as a little monograph to get my feet wet. For 33bucks that's what I expect and that's what I got, therefore the 33 bucks is a fair price.

Did the book show me how to build and train a net that makes money? Not really, as it is too basic. I read around in supplementary literature on technical trading, econometrics, statistics and last but not least chaos theory to finally get to the point of being able to coax a 14 to 20 day forecast out of net that is useful.

Those complaining about theory, academic snubs, sigma notation and the like...you are trying to use science and mathematics to produce a non-trivial result. To use another reviewer's remarks about another title, if you want to learn to compose / arrange great music you should not complain that the text you are consulting contain notes flat and sharp signs and other symbolism. If you are trying to learn to fly an aircraft you should not complain about having to learn about aerodynamics, navigation and meteorology. Why should it be any different with neural nets?

Get a good book. There are some around that only require a pre-requisite of college calculus and statistics and they really do provide an excellent intro into neural nets. You may want to try the following titles ( preferably in the order they are presented ):

1.) Neural Network Design ( Hagan, Beale, DeMuth)
2.) Neural Smithing ( Reeds and Marks )
3.) Neural Networks for Pattern Recognition ( Cristopher Bishop )

These titles are all available from Amazon.

You can read 1.) rather selectively ( it is well organized ) so is 2.). 3.) should be read front to back if time permits but it is rather self contained. By the way all three books combined will probably set you back a similar amount of money than Joe Zirrili's 135 pages.

Timothy Masters has a few titles on Neural Nets and Time Series prediction. Also for hands on experience try the neural network toolbox in Matlab.

Also check
http://www.faqs.org/faqs/ai-faq/neural-nets/part1/preamble.html
for further information and book tips.

Don't let some pompous snob with a Rolex talk you into spending 160 bucks in the hope that you will be able to harness neural network technology to predict stock prices. Developing neural nets tha tcan predict stocks is not easy.

If the above looks like too much effort, well then try writing a book with a similar title then pitch it for ...oh....shall we say ....50 to 100 bucks and try and sell it on the internet.

See how many fish will bite.

Maybe that will make you richer than predicting stocks with neural networks.


Regards

Dar Beh Dar

7 of 14 people found the following review helpful:

4.0 out of 5 stars Good Starting Point for Financial Types, February 13, 2000

By A Customer

This review is from: Financial Prediction Using Neural Networks (Hardcover)

Contrary to the academic snobs this book presents a basic overview with enough general content to get you going in this area. If you are looking for a seminal book on the subject full of derivatives and summation signs compounded with footnotes to obscure books whose seminal ideas are now dated, then look elsewhere. In the retrospect the book is now 3 years old. Given that technology and markets have changed the book still remains current. For you techies still having to buy this book maybe you never applied what the author also taught in the book. Maybe a little realization that this book is also geared to make you money should have any potential buyer realize that the stock market is made up of all kinds of folks who even with no education become millionaires. Such is the difference between Europe and America. Great starting point book to move on to next level with.