A Dynamic Linear Model Approach For Disaggregating Time Ser

An approach is presented for obtaining estimates of the basic (disaggregated) series, x subscript i, when only an aggregate series, y subscript i, of k period nonoverlapping sums of the disaggregated series is available. This is known as the disaggregation problem. It arises frequently in analyzing...

Full description

Saved in:
Bibliographic Details
Published in:Journal of forecasting 1989-04, Vol.8 (2), p.85
Main Author: Al-Osh, M
Format: Article
Language:English
Subjects:
Aggregations
Estimating techniques
Mathematical models
Time series
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An approach is presented for obtaining estimates of the basic (disaggregated) series, x subscript i, when only an aggregate series, y subscript i, of k period nonoverlapping sums of the disaggregated series is available. This is known as the disaggregation problem. It arises frequently in analyzing economic time series data. A dynamic linear model is set up for the problem and estimates of x subscript i are obtained by applying the Kalman filtering technique. An ad hoc procedure is developed for deriving a model form for the unobserved basic series from the observed model of the aggregates. An application of the proposed approach to a set of real data is given.
ISSN:0277-6693
1099-131X