Masters Theses


"The following studies the effectiveness of using fuzzy logic and neural networks for forecasting financial stock movements with technical indicators. A new Fuzzy-Neuro model for stock trading using technical analysis is also proposed and presented.

The main problem with technical analysis is choosing the right indicator and analyzing it. This study introduces two different models for combining different technical indicators. The first model is a fuzzy heuristic rule-based model. The second is a Fuzzy-Neuro model. The fuzzy rule-based model is able to combine two technical indicators (MACD and Williams’ %R) effectively, but making the rule base is difficult and human sentiments are induced into the system. For the second model the rule base is replaced by neural networks which are able to combine the three technical indicators (MACD, Williams’ %R, and RSI) to make effective trading decisions. The Fuzzy-Neuro model adds the capability of adjusting the weights of the three technical indicators that were used as inputs for the model. Since each indicator works better on different market conditions, adjusting the weights between the indicators produces a better return when a portfolio of stocks is traded. The Fuzzy-Neuro system also has the capability to learn from past data.

The proposed Fuzzy-Neuro model is tested on seven different companies actively traded in the U.S. markets. The proposed model shows promising results for a portfolio of seven companies"--Abstract, page 3.


Enke, David Lee, 1965-

Committee Member(s)

Dagli, Cihan H., 1949-
St. Clair, Daniel C.


Engineering Management and Systems Engineering

Degree Name

M.S. in Engineering Management

Research Center/Lab(s)

Intelligent Systems Center


University of Missouri--Rolla

Publication Date

Fall 2002


ix, 46 pages

Note about bibliography

Includes bibliographical references (pages 42-45).


© 2002 Vamsi Krishna Bogullu, All rights reserved.

Document Type

Thesis - Restricted Access

File Type




Subject Headings

Investment analysis
Stocks -- Prices
Fuzzy systems
Neural networks (Computer science)
Fuzzy logic

Thesis Number

T 8174

Print OCLC #


Link to Catalog Record

Electronic access to the full-text of this document is restricted to Missouri S&T users. Otherwise, request this publication directly from Missouri S&T Library or contact your local library.

Share My Thesis If you are the author of this work and would like to grant permission to make it openly accessible to all, please click the button above.