Time series analysis is an effective short-term forecasting. Stochastic process model is divided into two categories, autoregressive moving average process model and process model. The former is based on its lagged variable to calculate its future value, while the latter is the basis of past errors to calculate projected its future value.