Graduation Semester and Year
2011
Language
English
Document Type
Thesis
Degree Name
Master of Science in Information Systems
Department
Information Systems and Operations Management
First Advisor
Riyaz Sikora
Abstract
Momentum in financial markets can cause securities prices to continue trending upward/downward based on the recent performance. This paper reviews a study that attempts to discover how much daily returns in the stock market can be explained by financial momentum. This study uses classification data mining to attempt to predict the direction of daily returns of randomly selected stocks from the Russell 1000 and Russell 2000 stock indexes. The study uses moving averages of historical daily stock prices as attributes, along with different data mining classifiers, to attempt to make these predictions. A secondary goal of this study is to determine how effective using Distributed Data Mining (DDM) can be in predicting the direction of daily stock returns. Hence, DDM classifiers are used in the testing. This study discovers that the moving averages of daily returns do not help predict the direction of future daily stock returns any better than the percentages of returns from one trading day to the next. It also shows that the classifiers were no more than 60% accurate in predicting the directions of daily returns for any of the stocks used in this study. Hence, it appears that momentum cannot be used to explain very much of the movement in daily stock prices on a consistent basis.
Disciplines
Business | Management Information Systems
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
Evans, Stephen, "Data Mining In Financial Markets" (2011). Information Systems & Operations Management Theses. 8.
https://mavmatrix.uta.edu/infosystemsopmanage_theses/8