Predict Bankruptcy In Poland

From UCI Machine Learning Repository: SOURCE

Brief Introduction to Dataset

The dataset is about bankruptcy prediction of Polish companies.The bankrupt companies were analyzed in the period 2000-2012, while the still operating companies were evaluated from 2007 to 2013.

Overview:

The dataset consists of financial data used to predict bankruptcy among Polish companies. The data was collected from the Emerging Markets Information Service (EMIS, http://www.securities.com), focusing on the period from 2000 to 2013. The original dataset included five separate files, each corresponding to different forecasting periods. These periods reflect varying years of financial data and their corresponding bankruptcy status.

___________________________________________________________________________________________________________________

Original Files and Forecasting Periods

1stYear.arff: Contains financial data from the 1st year of the forecasting period with bankruptcy status after 5 years. It includes 7,027 instances (271 bankrupted, 6,756 non-bankrupted).

2ndYear.arff: Contains financial data from the 2nd year with bankruptcy status after 4 years. It includes 10,173 instances (400 bankrupted, 9,773 non-bankrupted).

3rdYear.arff: Contains financial data from the 3rd year with bankruptcy status after 3 years. It includes 10,503 instances (495 bankrupted, 10,008 non-bankrupted).

4thYear.arff: Contains financial data from the 4th year with bankruptcy status after 2 years. It includes 9,792 instances (515 bankrupted, 9,277 non-bankrupted).

5thYear.arff: Contains financial data from the 5th year with bankruptcy status after 1 year. It includes 5,910 instances (410 bankrupted, 5,500 non-bankrupted).