Rationalization of Exponential Smoothing in Terms of a Statistical Framework with Multiplicative Disturbances

It is established in this paper that exponential smoothing, in its most general linear form, is an optimal method of forecasting in large samples for time series with an irregular component, the size of which depends on a local mean. As such it is demonstrated that exponential smoothing has a statistical basis that extends beyond the framework of Box and Jenkins.


Issue Date:
Jul 01 1992
Publication Type:
Working or Discussion Paper
Language:
English
Total Pages:
15
Series Statement:
Working Paper No. 7/92




 Record created 2018-01-30, last modified 2018-01-31

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)