Novel feature selection methods to financial distress prediction

作者：

Highlights：

• An integrated feature selection method is proposed to predict distress firms.

• This method (HARC) embeds the experts’ knowledge with the wrapper method.

• The financial ratios are categorized into seven classes using experts’ knowledge.

• The prediction model based on HARC performs better than existing methods.

摘要

•An integrated feature selection method is proposed to predict distress firms.•This method (HARC) embeds the experts’ knowledge with the wrapper method.•The financial ratios are categorized into seven classes using experts’ knowledge.•The prediction model based on HARC performs better than existing methods.

论文关键词：Financial distress prediction,Genetic algorithm,Wrappers,Integrated prediction model,Feature selection

论文评审过程：Available online 19 October 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.09.047