Bankruptcy prediction is very important for all the organisations in the financial sector, as it affects the economy and leads to rise in many social problems with added incremental costs. There are large number of techniques that have been developed to predict the bankruptcy, which help the decision makers, such as investors and financial analysts. One of the bankruptcy prediction models is back-propagation neural network (BPNN). The paper first surveys the existing literature for various techniques that have been developed to assess credit risks, including the credit scoring models and quantitative models pioneered by Beaver and Altman, which focuses on the borrower's inability to meet credit obligation. Thereafter, this paper, using the tailored BPNN, endeavours to predict the financial ratios expressing the position of a firm to regulate the bankruptcy and assess the credit risks. It first estimates the financial ratio for a firm from 2001 to 2008 to the train the BPNN and uses the estimates of the year 2009 and 2010 values in the validation process. Finally, it dwells to draw predictions for the period 2011—2015 and emphasises the growing role of BPNN application-based prediction models for banking sector with a case study of Oriental Bank of Commerce. We conclude with practical suggestions on how best to integrate models and research into credit lending decisions.
Credit lending, Back propagation neural network, Credit lending, Ratio analysis
