Resource title

Arma models and the box jenkins mehodology

Resource image

image for OpenScout resource :: Arma models and the box jenkins mehodology

Resource description

This paper studies the Box-Jenkins methodology to ARMA models and determines the reasons why in empirical tests it is found that the post-sample forecasting accuracy of such models is worse than much simpler time series methods. We conclude that the major problem is the way of making the series stationary in its mean (the method of differencing) that has been proposed by Box and Jenkins. If alternative approaches are utilized to remove and extrapolate the trend in the data, ARMA models outperform the corresponding methods involved in the majority of cases. It is shown that using ARMA models to seasonally adjusted data improves post-sample accuracies while simplifying the use of ARMA models. It is also confirmed that transformations slightly improve post-sample forecasting accuracy, particularly for long forcasting horizons. Finally, it is demonstrated that AR(1) and AR(2), or their combination, produce as accurate post-sample results as those found through the application of the Box-Jenkins methodology

Resource author

Resource publisher

Resource publish date

Resource language


Resource content type


Resource resource URL

Resource license

Copyright INSEAD. All rights reserved