ARIMA Modelling and Forecasting of Time Series (Course code 0433)


ARIMA stands for AutoRegressive Integrated Moving Average. ARIMA intervention models are used to describe and forecast time series. ARIMA forecasts may be useful as substitutes for preliminary data that are not yet available or in determining the credibility of such data. Intervention models can be used to measure the impact that events such as shifts in level, outliers and the introduction of new taxes or regulations have on the series being studied. Transfer functions are used to relate one series not only to its past values but also to present and past values of other time series.

Benefits to participants

At the end of the training period, participants will be able to describe and explain the required theory and use the PC SAS statistical system to fit ARIMA models, intervention models or transfer function models to time series.

Target population

This course is intended for employees who have a solid grounding in statistics (i.e. they understand the terms random variable, mean, variance, covariance and linear regression).

Course outline

  • The steps in fitting an ARIMA model are the identification of the model, the estimation of the model's parameters, diagnostic checking of the model and forecasting of future observations.
  • The operations and the theoretical concepts needed to perform these steps.


Familiarity with Windows and the ability to work with software for statistical analysis on a personal computer.

Duration: 5 days