Designing Stock Market Trading Systems: With and Without Soft Computing by Bruce Vanstone & Tobias Hahn

Author:Bruce Vanstone & Tobias Hahn [Vanstone, Bruce & Hahn, Tobias]
Language: eng
Format: epub
Tags: Business & Economics, E-Commerce, Online Trading
ISBN: 9780857191359
Google: g_w2o6ldgdIC
Publisher: Harriman House Limited
Published: 2010-08-23T00:09:02.988525+00:00

Consider the case of a simple moving average crossover system. Such a system will invariably be subject to whipsaw. This is where a signal occurs, but is quickly reversed, maybe even in the next time period. This causes the system to move in and out of trades, tying up capital, and incurring transaction costs. The typical trader’s solution is to implement some crossover threshold which must be exceeded before the trade is taken, but this delays entry, and in the case of a profitable outcome, costs time and money.

A better solution might be to look to a neural network to help solve this problem. An ANN is a universal approximator; it is capable of determining a non-linear relationship between a set of inputs and a set of outputs.

The main challenge is to find a way to express your problem so that an ANN can help provide the solution. Once this is done, the process of creating the required ANN can begin.

The main functions in building an ANN are:

Expressing the problem to be solved in such a way that an ANN can be used to build the solution

Partitioning data into training and testing sets

Finding variables of influence

Making ANN architecture choices

Training

Testing

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.