Optimal spatial aggregation of space–time models and applications

Cancers are serious health concerns for every country. In the U.S. various cancer data are collected, monitored, and studied by the American Cancer Society (ACS). Since the data involves both spatial and temporal components, space–time models are useful for their analyses. Often these data (such as...

Full description

Saved in:
Bibliographic Details
Published in:Computational statistics & data analysis 2020-05, Vol.145, p.106913, Article 106913
Main Authors: Gehman, Andrew, Wei, William W.S.
Format: Article
Language:English
Subjects:
Aggregation algorithm
GSTARMA model
Optimality
STARMA model
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Cancers are serious health concerns for every country. In the U.S. various cancer data are collected, monitored, and studied by the American Cancer Society (ACS). Since the data involves both spatial and temporal components, space–time models are useful for their analyses. Often these data (such as cancer rates) from varying geographical or political areas will be aggregated spatially to correspond to larger regions for analysis at that spatial scale. Methods to compare spatial aggregation schemes and to identify the optimal spatial aggregation are introduced. Specifically, some useful theorems and algorithms to determine the aggregation scheme that results in the minimum aggregate model error will be given, and they are demonstrated using the U.S. ovarian cancer incidence.
ISSN:0167-9473
1872-7352
DOI:10.1016/j.csda.2020.106913