Forecasting Iran ’ s Rice Imports Trend During 2009 - 2013

I n the present study Iran's rice imports trend is forecasted, using artificial neural networks and econometric methods, during 2009 to 2013, and their results are compared. The results showed that feet forward neural network leading with less forecast error and had better performance in comparison to econometric techniques and also, other methods of neural networks, such as Recurrent networks and Multilayer perceptron networks. Moreover, the results showed that the amount of rice import has ascending growth rate in 2009-2013 and maximum growth occurs in 2009-2010 years, which was equal to 25.72 percent. Increasing rice import caused a lot of exchange to exit out of the country and also, irreparable damage in domestic production, both in terms of price and quantity. Considering mentioned conditions, economic policy makers should seek ways to reduce increasing trend of rice import; and more investment and planning for domestic rice producers. INTRODUCTION Rice is considered as the one of the most important cereal in human consumption (Kazem-nejhad & Mehrabi Boshrabadi, 1999, p. 104). Among cereals, rice is the only product that is grown for human consumption. Rice, after wheat, is the second most consumed crop in Iran that its use has been increased since 50s, significantly (Noori, 2002, p. 27). At present, Iran is considered as one of the major importers of rice. During 2008, nearly 1383 thousand tons of rice, worth 790 million dollars, was import to the country that compared to last year, was increased about 30 percents, in term of weight, and 85 percents, in term of value. Thus, proper policies are necessary to reduce import and increase domestic production of rice, that this would not be possible regardless of the future status of the country's imports. Today, the importance of economic variables forecasting is obvious for economic programmers and policy makers. Therefore, in the recent decades, various models of forecasting were developed and also competed with each other. In fact, in order to evaluate the efficiency of this model, the accuracy of their predictions should be compared. Thus, a model will be successful in explaining the relationship between variables if it is able to forecast the future values of variables accurately. In the past, economists were trying to explain the current situation and forecasting the future values of dependent variables as well as providing economic policies, by using various econometric methods. In mid 90s, these important deficiencies and importance of forecasting …