2 results
Search Results
Now showing 1 - 2 of 2
Article Citation - WoS: 2Predicting Financial Failure of the Turkish Banks(World Scientific Publ Co Pte Ltd, 2006) Doganay, M. Mete; Ceylan, Nildag Basak; Aktas, RamazanBanks are the most important financial institutions in Turkey because other financial institutions are not developed efficiently yet. Turkish banks experienced financial difficulties and a substantial amount of banks failed in the past. This event urged the government to initiate measures to prevent banks from getting into financial difficulties. As a result of these measures, Turkish banking system currently seems to be very attractive for the foreign investors willing to invest in this sector. One of the main concerns of the foreign investors is a possibility of a new banking crisis although it is very remote at this time. The purpose of this study is to develop early warning systems predicting the financial failure at least three years ahead of financial date. A number of multivariate statistical models such as multiple regression, discriminant analysis, logit, probit are used. We found that the most appropriate model is logit. The significant variables obtained from the models explain very well the causes of the bank failures. Our models can be used to assist interested parties to predict the probability of financial failure of Turkish banks.Article Citation - WoS: 19Citation - Scopus: 24Increasing Accuracy of Two-Class Pattern Recognition With Enhanced Fuzzy Functions(Pergamon-elsevier Science Ltd, 2009) Celikyilmaz, Asli; Tuerksen, I. Burhan; Aktas, Ramazan; Doganay, M. Mete; Ceylan, N. BasakIn building an approximate fuzzy classifier system, significant effort is laid oil estimation and fine tuning of fuzzy sets. However, in such systems little thought is given to the way in which membership functions are combined within fuzzy rules. In this paper, a robust method, improved fuzzy classifier functions (IFCF) design is proposed for two-class pattern recognition problems. A supervised hybrid improved fuzzy Clustering for classification (IFC-C) algorithm is implemented for structure identification. IFC-C algorithm is based oil it dual optimization method, which yields simultaneous estimates of the parameters of (c-classification functions together with fuzzy c partitioning of dataset based oil a distance measure. The merit of novel IFCF is that the information oil natural grouping of data samples i.e., the membership values, are utilized as additional predictors of each fuzzy classifier function to improve accuracy of system model. Improved fuzzy classifier functions are approximated using statistical and soft computing approaches. A new semi-non-parametric inference mechanism is implemented for reasoning. The experimental results Of the new modeling approach indicate that the new IFCF is it promising method for two-class pattern recognition problems. (c) 2007 Elsevier Ltd. All rights reserved.
