ARIMA is one of the most versatile linear models for forecasting seasonal time series. It has enjoyed great success in both academic research and industrial applications during the last three decades.The class of ARIMA models is broad. It can represent many different types of stochastic seasonal and nonseasonal time series such as pure autoregressive (AR), pure moving average (MA), and mixed AR and MA processes. The theory of ARIMA models has been developed by many researchers and its wide application was due to the work by Box et al.