A Computational Approach for Selecting the Performing Stocks of the Dow
Abstract
This paper proposes a computational approach that integrates the generalized regression neural network (GRNN) and two supporting technologies, namely information gain and principal component analysis, to predict stock return performance. The integrated approach takes advantage of the synergy among these technologies to collect the relevant data from various financial variables, transform the data so that they will be uncorrelated, and estimate the performance of stock return from the underlying data. Thirty stocks listed in Dow Jones Industrial Average Index (DJIA) were selected in this research. In a simulation with out-of-sample data, trading is performed when the stocks achieve the estimated future returns of at least ±2% performance, representing a reasonable opportunity to take a position. The results show that the average trading profits guided by the GRNNs are 17.52% and 17.21% higher than those obtained from investing in the DJIA index and the buy-and-hold accounts, respectively.
Recommended Citation
S. Thawornwong and D. L. Enke, "A Computational Approach for Selecting the Performing Stocks of the Dow," Intelligent Engineering Systems Through Artificial Neural Networks, vol. 12, pp. 695 - 700, American Society of Mechanical Engineers, Dec 2002.
Department(s)
Engineering Management and Systems Engineering
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 American Society of Mechanical Engineers, All rights reserved.
Publication Date
01 Dec 2002
