Forecasting quarterly profit growth rate using an integrated classifier

Abstract

This study proposes an integrated procedure based on four components: experiential knowledge, feature selection method, rule filter, and rough set theory for forecasting quarterly profit growth rate (PGR) in the financial industry. To evaluate the proposed procedure, a called PGR dataset collected from Taiwan's stock market in the financial holding industry is employed. The experimental results indicate that the proposed procedure surpasses the listing methods in terms of both higher accuracy and fewer attributes. © 2011 Springer-Verlag Berlin Heidelberg.