پيش بيني قيمت محصولات كشاورزي: مطالعه مورد پياز ، سيب زميني و گوجه فرنگي

Forecasting the Price of A~r icultural Products: A Case Study of Onions, Potatoes, and Tomatoes

This study is curried out with the objective or forecasting nominal and real prices or some L1gricultural products such as onions. potatoes and tomatoes. Alter the analysis or stationarity or the associated limeseries. the randomness of the variables is studied using the nonparametric Wald-Wullowirz and the parametric Durbin-Watson tests, lhe results show thaL all the nominal price series or the said products and the rCJI price series 01" potatoes were identified as non-random and forccastable time-series. The time-horizon for this study is 135013X4. Models used lor the I(JlcGlSting in this analysis were AR, M/\, ARIMA. Single l-xponcntia] Smoothing, Double Exponential Smoothing, l larrnonic Analysis. ARelI and the Artificial Ncuronetwork. Amongst the four forecasting series based on the minimumerror criterion, the AR and MA models forecasted the normnal price series or onions and potatoes much better. The Harmonic model forecasted the !10111 Ina I pncc or tomatoes with minimal error. The forecasted values Ior the real price of potatoes using the ARCH model produced the nunimum errors. Forecasting error for the nominal price series of potatoes \\ a:-. less than its real price series. The forecasting values (or the year 13X5 and 13X6 were shown to he mere than the similar values for 1384, with the exception of the price for potatoes.