Forecasting sugar beet price in Iran

Agricultural prices have a high fluctuation and forecasting may help decision making effectively. The aim of this study was to forecast the nominal and real prices of sugar beet and to recognize the appropriate forecasting model. Initially the stationary of the series was tested. In order to investigate whether the series are stochastic, the nonparametric test of Vald-Wulfowitz and parametric test of Durbin-Watson were then applied. The results indicated that the nominal price of sugar beet was non stochastic and predictable while the real price series were found to be stochastic. The study period covered 1971- 2005. The Autoregressive, Moving Average, ARIMA, Single and Double Exponential Smoothing, Harmonic and ARCH were applied to forecasting sugar beet prices. Based on the lowest forecasting error criterion, the Harmonic model was selected as the best model. The prices of sugar beet forecasted by different models were at the range of 344000-396000 and 398000-448504 Rials per ton for 2004 and 2005, respectively. The sugar beet actual price for 2004 and 2005 were 387200 and 447000 Rials per ton, respectively.