Financial distress prediction based on ensemble classifiers of multiple reductions

Financial distress prediction is an important research topic in both academic and practical world. This paper puts forward a financial distress prediction model based on multiple reduction ensembles, which employs neighborhood rough set based attribute reduction to generate a set of reducts, then each reduct is used to train a base classifier, and finally their results are combined through simple majority voting. Taking Chinese listed companies´ real world data as sample data, adopting 10-fold cross validation technique to assess predictive performance, an experiment study is carried out. By comparing the experiment results with the raw data and the single reduct based classifiers, it is concluded that this model can improve the average prediction accuracy or both accuracy and stability, so it is more suitable for financial distress prediction than the single reduct based classifiers.