Modeling the Seasonal Behavior of Iran's GDP with Emphasis on Agriculture Sector: Composition by Sectors

This study tries to model the seasonal behavior of GDP in different economic sectors (agriculture, services, oil) using periodic autoregressive (PAR) and seasonal integration (SI) models during 1998:3-2010:6. According to the results, the GDP of agricultural sector has a regular and periodic behavior, therefore employing the periodic autoregressive model for GDP behavior of this sector can be very effective. Results of the Hylleberg et al (1990) seasonal unit root test showed seasonal behavior in service sector, so the data became stationary using appropriate filters and then the appropriate seasonal integration model was estimated. The oil sector showed no seasonal behavior, and autoregressive integrated moving average (ARIMA) model is applied to model the GDP of this sector. Finally, the fitted models applicated to forecast the next two years production in economic sectors. So, due to the different nature of the various economic sectors, studying the sectors independently is recommended