Ensembles of Local Linear Models for Bankruptcy Analysis and Prediction

Résumé

Bankruptcy prediction is an extensively researched topic. Ensemble methodology has been applied to it in past work. However, the interpretability of the results, so often important in practical applications, has not been emphasized. This paper builds ensembles of locally linear models using a forward variable selection technique. The method applied to four datasets provides information about the importance of the variables, thus offering interpretation possibilities.