İşletme Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/403
Browse
2 results
Search Results
Article Towards Predicting Financial Information Manipulation(2007) Aktaş, Ramazan; Alp, Ali; Doğanay, Mehmet MeteManipulation 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.Article Citation - WoS: 20Citation - Scopus: 28Prediction of Financial Information Manipulation by Using Support Vector Machine and Probabilistic Neural Network(Pergamon-elsevier Science Ltd, 2009) Ogut, Hulisi; Aktas, Ramazan; Alp, Ali; Doganay, M. MeteDifferent methods have been used to predict financial information manipulation that can be defined as the distortion of the information in the financial statements. The purpose of this paper is to predict financial information manipulation by using support vector machine (SVM) and probabilistic neural network (PNN). A number of financial ratios are used as explanatory variables. Test performance of classification accuracy, sensitivity and specificity statistics for PNN and SVM are compared with the results of discriminant analysis, logistics regression (logit), and probit classifiers, which have been used in other studies. We have found that the performance of SVM and PNN are higher than that of the other classifiers analyzed before. Thus, both classifiers can be used as automated decision support system for the detection of financial information manipulation. (C) 2008 Elsevier Ltd. All rights reserved.