SSE/EFI Working Paper Series in Economics and Finance
Michael K. Andersson
Do Long-Memory Models Have Long Memory?
Abstract: This paper examines the predictability memory of
fractionally integrated ARMA processes. Very long memory is found for
positively fractionally integrated processes with large positive AR
parameters. However, negative AR parameters absorb, to a great extent, the
memory generated by a positive fractional difference. An MA parameter may
also reduce the predictability memory substantially, even if the parameter
alone provides hardly any memory.
Keywords: ARMA; Fractional integration; Prediction horizon; (follow links to similar papers)
JEL-Codes: C22; C52; C53; (follow links to similar papers)
4 pages, February 27, 1998, Revised March 16, 2000
- This paper is published as:
Andersson, Michael K., (2000), 'Do Long-Memory Models Have Long Memory?', International Journal of Forecasting, Vol. 16, Jan-March, No. 1, pages 121-124
Questions (including download problems) about the papers in this series should be directed to Helena Lundin ()
Report other problems with accessing this service to Sune Karlsson ()
or Helena Lundin ().
Design by Joachim Ekebom