Corporate Bankruptcy Prediction using Machine Learning
The study compares and constrasts several supervised learning algorithms to investigate the accuracy of predicting bankruptcy
The prediction of corporate bankruptcy is an important and widely studied topic (Wilson and Sharda, 1994). Creditors and investors need to be able to predict the probability of default for profitable business decisions. For banks, accurate assessment of the probability of bankruptcy can lead to more profitable lending practices as well as better estimates of interest rates that reflect credit risks. Bankruptcy prediction has been a popular subject for business researchers
In this study I try to investigate the accuracy of predicting bankruptcy using methods such as :
Logistic Regression
Tree based methods
General Additive Models
Neural Networks