Normal view MARC view

Arma models and the box jenkins mehodology

Author: Makridakis, Spyros ; Hibon, MichèleINSEAD Area: Technology and Operations Management Series: Working Paper ; 95/33/TM Publisher: Fontainebleau : INSEAD, 1995.Language: EnglishDescription: 17 p.Type of document: INSEAD Working Paper Online Access: Click here Abstract: 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 Next title: Arma models and the box jenkins mehodology (RV of 95/33/TM) - Hibon, Michèle;Makridakis, Spyros - 1995 - INSEAD Working Paper
Tags: No tags from this library for this title. Add tag(s)
Log in to add tags.
Item type Current location Collection Call number Status Date due
INSEAD Working Paper Digital Library
PDF Available

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

Digitized

There are no comments for this item.

Log in to your account to post a comment.
Koha 3.18 - INSEAD Library Catalogue
Library Home | Contact Us | What's Koha?