Online Program

Return to main conference page
Friday, February 15
Fri, Feb 15, 5:15 PM - 6:30 PM
St. James Ballroom
Poster Session 2 and Refreshments

Dimension Reduction in Bankruptcy Prediction: A Case Study of North-American Companies (303834)

View Presentation View Presentation

*Edward Alexander Golas, Bryant University 
Son Nguyen, Bryant University 

Keywords: Bankruptcy Prediction, Solvency Prediction, Imbalanced Data, Rare Events, Classification, Sliced Inversed Regression

Due to its significance in the world of accounting, finance and investment, bankruptcy prediction has attracted a great deal of research in the data mining/machine learning community. This poster examines the influence of different dimension reduction techniques on machine learning models applied to the bankruptcy prediction problem. The studied dimension reduction techniques include Principal Component Analysis (PCA), Sliced Inverse Regression (SIR), Sliced Average Variance Estimation (SAVE) and Factor Analysis (FA). To best isolate the effects of dimension reduction techniques on the predictive power of a model only one modeling method was used, decision tree, and all data was balanced using under-sampling. Our computation shows that dimension reduction techniques can have a great affect the performance of predictive models and that one could use dimension reduction techniques to improve the predictive power of the decision tree model. Also, in this study, we propose a method to estimate the true dimension of the data.