摘要：This paper used the X-12 method, seasonal unit root and Autoregressive Integrated Moving Average Model to identify and to model the generator process of rural employment in the state of São Paulo in the period from January 1996 to December 2006. The results show that there is strong seasonal demand for occasional rural employment from April to August for several main harvests. The seasonal unit root test detected the presence of seasonal unit root. This result confirms the presence of a seasonal pattern in the rural employment time series. The ARIMA model caught the rural employment dynamism for occasional workers. It was necessary to add three moving average parameters of small orders, beyond a seasonal moving average parameter, and an autoregressive parameter which represents the hiring of permanent workers.
