A Hybrid RST and GA-BP Model for Chinese Listed Company Bankruptcy Prediction

Bankruptcy is a worldwide economic and social problem with high costs. And the high social costs associated with bankruptcy have spurred searches for better theoretical understanding and prediction capability. In this paper, we suggest a hybrid approach to Chinese listed company bankruptcy prediction, using a GA-BP (genetic algorithm and back propagation) model to construct a bankruptcy prediction model with variables derived by rough set theory (RST). An example is given to validate this model. The results show our hybrid model has higher prediction accuracy and less execution time in bankruptcy prediction when compared against GA-BP algorithm.