روشي مركب از اقتصادسنجي و شبكه هاي عصبي مصنوعي براي پيش بيني پديده هاي اقتصادي

A HYBRID METHOD OF ECONOMETRICS AND ARTIFICIAL NEURAL NETWORKS FOR FORECASTING: A CASE STUDY

Precise and reliable prediction of economical phenom­ena is one of the most important goals of economists. Economists have used econometric methods to predict and analyze economic phenomena for several years. Re­cently, methods, such as Neural Networks, Fuzzy Logic and Genetic Algorithms, have been introduced to economists as new methods for the prediction of econom­ical phenomena, lu this paper, based on a combination of Econometric and Neural Network approaches, a new method is introduced, which is more precise and more reliable than the typical Econometric or Neural Network methods. Its reliability has come from its Econometric function and its precision has come from its Neural Net­work part. In order to observe the performance of the new method, the number of demands of financial facili­ties from one of the development banks of Iran are mod­elled and the results compared with those of the typical methods. The results of the comparison show the effi­ciency of the proposed method.