Methods for choice of model in descriptive time se-ries : A review with example

This paper is a review of existing methods with illustrative examples for choice of model in descriptive time series. The methods reviewed are three graphical methods, X-12 ARIMA method and the method of seasonal differences and quotients. For each method, the same illustrative example and simulated data were used to demonstrate each method. The results obtained from the various methods using simulated additive and multiplicative series show that the various methods identified correctly the appropriate model for decomposition. Also, applying the various methods to quarterly sales of petroleum products from 2004-2013, the result reviewed that the appropriated model for decomposition of the series is the multiplicative model.