A Stock Selective System by Using Hybrid Models of Classification
Stock trade is a popular investing activity and during this activity, investors expect to gain higher profit with lower risk. Therefore, the problem of predicting stock returns has been an important issue for many years. This study is aimed on the discover relationship between financial data of public companies and return on investment by using data mining technology. The study propose a stock selective system by using hybrid models of classification. Use the hybrid models of association rules, cluster, and decision tree, it can provide meaningful decision rules for stock selection for intermediate- or long-term investors. Further, these rules are use to select some profitable stocks of the following years. The outcome evidences the higher return on investment in proposed model than general market average.
KeywordsFinancial indexes Relation rules Cluster Decision Tree
Unable to display preview. Download preview PDF.