IJPAM: Volume 58, No. 2 (2010)

A COMPARISON OF ESTIMATION METHODS FOR
CONDITIONALLY-SPECIFIED SPATIAL DATA

Monica C. Jackson
Department of Mathematics and Statistics
American University
4400, Massachusetts Ave., Washington, DC, 20016, USA
e-mail: [email protected]


Abstract.The objective of a spatial data analysis may be to capture the trends apparent in the data set. In spatial data, we do not have the unidirectional flow of time that occurs with time series data. Instead, spatial models are built on nearest neighbors. For discrete data, there are several models including generalized linear mixed models and conditionally specified models such as the auto-Poisson model. Each has its drawbacks. We evaluate existing methods that model the distribution of discrete outcomes when the value at a location depends on its neighbors under certain practical problems of edge data, large number of zeros, and outliers.

Received: December 15, 2009

AMS Subject Classification: 62H11

Key Words and Phrases: spatial statistics; spatial correlations

Source: International Journal of Pure and Applied Mathematics
ISSN: 1311-8080
Year: 2010
Volume: 58
Issue: 2