Towards Predicting Financial Information Manipulation

Abstract

Manipulation is one of the important issues in securities markets because manipulative actions send false signals to the investors and make them buy or sell securities they otherwise would not buy or sell. There are different types of manipulations that can deceive investors. One type of manipulation is financial information manipulation. Manipulators, who use this type of manipulation, distort information in the financial statements in order to give false information about the prospects of the issuing firms. This paper attempts to predict financial information manipulation by using the multivariate statistical techniques and neural networks. A number of financial ratios are used as explanatory variables. The multivariate statistical techniques used are discriminant analysis, logistics regression (logit), and probit. Unlike other studies, the present study takes multicollinearity between financial ratios into account and conclude that the estimated multivariate statistical models rather than the neural networks can be used as early warning systems to detect possible financial information manipulations.