A Bankruptcy Prediction Model Using Random Forest

A Bankruptcy Prediction Model Using Random Forest

Ruchi Gupta1,Nikhil Patil

ASM’s Institute of Management and Computer Studies

Mumbai University

Mumbai , India

Email: 1 guptaruchi1206@gmail.com, 2nikhilppatil678@gmail.com

Abstract—Bankruptcy Prediction is very important for an  organization as well as decision makers such as financiers and  investors. Selection of dataset for training the prediction model,  the Machine Learning tool used for prediction and various other  factors are essential in building an efficient prediction model. The  dataset includes financial ratios as attributes that are derived from the financial statements of various companies. The most  influencing ratios that are required for predicting bankruptcy are  selected on the basis of the Genetic Algorithm which filters out the  most important ones from different existing bankruptcy models. These ratios of different companies are fed as an input to train the  model being implemented in R. The prediction algorithm used is  Random Forest, which will enable us to differentiate between  bankrupt and non-bankrupt companies.

Keywords—bankruptcy prediction, bankruptcy models,  financial ratios, genetic algorithm, random forest, decision  trees, RStudio