Study of financial data of listed Indian companies to predict fraud

Abstract

The study during this project is to analyze financial data of fraudulent and non-fraudulent companies in India and derive a model that will help to predict fraud. Quantitative methods such as the Beneish Model (Beneish M, 1999) is primarily used to analyze financial data and assess the accuracy of the fraud prediction for Indian companies. The beneish model uses eight financial indices to determine whether a company has manipulated its earnings and provides an M-score to describe the degree of manipulation. The new variables such as inventory and revenue growth are introduced additionally into the model to assess whether overall accuracy of the model improves. Further, other classification models such as decision trees and random forests are also used to compare the results for consistency.