Abstract
A sampling distribution is the frequency distribution of some statistic constructed by taking all possible samples of a given size from a parent population. Sampling distributions of general interest most often are for model parameter estimates, means, variances, and correlation coefficients. All of these distributions are affected by nonzero spatial autocorrelation. Sampling distributions of particular interest in spatial analysis are those for MC, GR, and ρ̂, the autocorrelation parameter of a spatial autoregressive model. The key to establishing a sampling distribution is stipulating what constitutes a sample. Sampling distributions can be explored through use of simulation techniques, resampling procedures, and algebraic analysis.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
A general discussion of MCMC appears in Gilks et al. (1996) and includes a disease mapping case study. Geographic data simulation of this type is outlined in Heagerty and Lele (1998).
The Gibbs sampler is explained by Casella and George (1992) and is demonstrated in terms of an auto-exponential and Ising models by Robert and Casella (1999, pp. 286–287).
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Griffith, D.A. (2003). Sampling Distributions Associated with Spatial Autocorrelation. In: Spatial Autocorrelation and Spatial Filtering. Advances in Spatial Science. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24806-4_3
Download citation
DOI: https://doi.org/10.1007/978-3-540-24806-4_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-05666-6
Online ISBN: 978-3-540-24806-4
eBook Packages: Springer Book Archive