Three approaches to modelling spatial data in which simulation plays a vital role are described and illustrated with examples. The ยฎrst approach uses ยฏexible regression models, such as generalized additive models, together with locational covariates to ยฎt a surface to spatial data. We show how the b
Simulation of spatially correlated data in two dimensions
โ Scribed by P.I. Brooker
- Publisher
- Elsevier Science
- Year
- 1985
- Tongue
- English
- Weight
- 191 KB
- Volume
- 27
- Category
- Article
- ISSN
- 0378-4754
No coin nor oath required. For personal study only.
โฆ Synopsis
Matheron's turning bands method for the simulation of spatially correlated data in two or three dimensions requires that the relationship between the covariance obeyed by the realizations first generated on lines, and the covariance of the two or three dimensional process must be solved. In three dimensions the solution is immediate, but in two dimensions the integral equation relating the covariances of the one and two dimensional processes is more complex. This equation was developed and solved numerically by Brooker and Paul (1982). In this paper the analytic solution is presented and applied to the most con-unon model used in ore deposit description.
๐ SIMILAR VOLUMES
In the context of the first-order auto-logistic model for dichotomous spatial data, we obtain the first statistical test for comparing two geographical areas on the basis of their statistical properties. These optimal unbiased tests are illustrated with an example which features the Markov chain Mon
## Abstract A random flight model of linear transport processes in two spatial dimensions is considered and solved exactly in closed algebraic form. Its oneโdimensional version had been proposed by Taitel as a means to overcome the paradox of infinite speed of propagation within classical heat diff