Bankruptcy Prediction: Dynamic Geometric Genetic Programming (DGGP) Approach مقاله

"Results of Dynamic Geometric Genetic Programming (DGGP) classification methodology is compared with ∗ Department of Mathematics, University of Semnan, Iran & Risk Management Unit, Pasargad Bank, Tehran, Iran ∗∗ Monetary and Banking Research Institute, Central Bank of the Islamic Republic of Iran, Tehran, Iran common and transformed ratios. In this paper we construct a new methodological approach to financial ratio analysis called the Dynamic Risk Space measure (DRS) which involves data transformation and we illustrate the use of this method for measuring financial risk in prediction of bankruptcy risks. As may be observed from the prediction results in Section 4, we suggest the use of Dynamic Geometric Genetic Programming (DGGP) methodology for financial variables analysis, which provide a conceptual and complimentary methodological solution to many problems associated with the use of common ratios. (2008) we tested these selected variables with Genetic Programming (GP) to obtain fitness function tree and to illustrate that this new transformation will predict more accurate and can be used as an alternative for common ratios. Due to better performance testing of this new transformation, data set collected do not have to be in a particular industry type or similar firm size and application of outlier deletion method to overcome any potential explanatory effect errors that will be caused by independent variables distribution1 is not necessary to accomplish the new model properties as explained in methodology. We found a rise in classification accuracy with the application of this new independent variables transformation using genetic programming (GP) technique as a statistical prediction model compared to input common ratios as independent variables."