An improved rough set data model for stock market prediction

Rough set theory is a well established tool for dealing with inconsistent data. The dependencies among the attributes, their significance, and evaluation can easily be performed using intelligent data analysis tool viz., rough set theory. The objective of this article is to modify the existing stock market predictive model based on rough set approach by A.E Hassanien et al. and to construct a data model that would generate fewer number of decision rules. Moreover the results obtained from the proposed data model are compared with well-known software tool Rough set Exploration system 2.2 popularly known as RSES 2.2. It is shown that the proposed model has a higher overall accuracy rate and generates more compact and fewer rules than RSES 2.2. Rough confusion matrix is used to evaluate the predicted classification performances. The effectiveness of this data model is demonstrated on data set consisting of daily movements of a stock traded in Kuwait Stock Exchange spanning over a period of five years.