Designing of Hybrid Early Warning Model of Financial Crisis in Iran's Economy

Document Type : Research Article


1 Faculty of Islamic Studies and Management Imam Sadeq(as) University

2 Academic Member, Faculty of Economics, University of Tehran

3 Associate Professor, Faculty of Management, University of Tehran


From a long time ago, predicting of financial crisis has been one of the most important concerns of economists and financial economics experts. In the first section of the present research, based on analyzing the fluctuation of rate of change of 10 Macro.variables during the period of 1358-1391 in Iran's economy using Static and Dynamic, Variance and Semi Variance, and based on "Self-Organizing Map" neural network model, the exact years of financial crisis in Iran' economy have been determined. Through the second section, using the "Feed Forward" neural network model, the future value of the rate of change of Macro.variables during 1392-1395 in Iran's economy have been predicted, and finally via the third section using the "Pattern Net" neural network model and based on results of two previous sections, occurrence of financial crisis in the years of 1392-1395 has been foretold. Based on final results of the present research, it can be indicated that the financial crisis which has been occurred in the year of 1391 in Iran' economy, which is the consequence of occurrence of financial crisis in1389 and 1390, will be continued in to the year of 1392; however, it will be evaporated gradually during it and the years of 1393 and 1395 are the years without financial crisis; albeit the model is indicating some early warning signals due to the return of crisis to Iran’s economy at the year of 1394.