Seasonal forecasting of meat prices in Iran: Application of periodic autoregressive model

Meat has always been one of the most important commodities for Iranian households, and most of the times have had the largest households’ budget share among the goods and beverages groups. This study aims to estimate and forecast the seasonal price of meats in Iran. To address this aim, we used seasonal data of chicken, beef and lamb for the period of 1998-2011. In this context, first HEGY test was applied to check the seasonal unit root, then Fanses & Paap’s periodic unit root test and Boswijk & Franses’s seasonal periodic behavior test were used. Results indicated that periodic autoregressive model of order one [PAR(1)] is the best fit for chicken price forecasts. The seasonal unit root test for beef and lamb also gives that they follow autoregressive moving average patterns; therefore ARIMA model is a suitable model to forecast the prices of these commodities.