Forecasting Prices of Chicken and Egg by Using Artificial Neural Network in Iran

The aim of this study was to choose forecasting model for nominal and real price of beef, sheep meat, Milk, and Wool. Initially the stationary of the series was tested, then in order to investigate whether series are stochastic, nonparametric test of Vald-Wulfowitz was applied. The study period covers 1346-1384. Based on the above tests results, all of the selected nominal and real prices were recognized to be predictable. The models applied to forecast are ARIMA, and Artificial Neural Network (ANN). The findings indicated the relative superiority of ARIMA in comparison with ANN in predicting nominal prices of selected products. However, In the case of real prices ANN showed a comparativesuperiority. It was also found that in the case of nominal series increase in the forecast period lead to increased forecast error.