Local spatial heteroscedasticity (LOSH)
- 396 Downloads
In keeping with the Annals of Regional Science 50-year tradition of emphasizing spatial analytic contributions, a new statistic, H i , is introduced as an extension of the recent work on map pattern analysis using local spatial criteria. In conjunction with local statistics for the mean level of a spatial process, H i tests for local spatial heteroscedasticity. The statistic measures spatial variability while attempting to avoid the pitfalls of using the non-spatial F test in georeferenced situations. H i is defined and suitably scaled, and an inferential procedure is given. Three artificial cases are explored: a random pattern, a cluster pattern of high values, and a binary-based cluster. The statistic may be used to identify boundaries of clusters and to describe the nature of heteroscedasticity within clusters.
JEL ClassificationC21 C46 C51
Unable to display preview. Download preview PDF.
- Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57(1): 289–300Google Scholar
- Box GEP (1953) Non-normality and tests on variances. Biometrika 40: 318–335Google Scholar
- Cliff AD, Ord JK (1969) The problem of spatial autocorrelation. In: Scott AJ (ed) London papers in regional science. Pion, London, pp 25–55Google Scholar
- Cliff AD, Ord JK (1973) Spatial autocorrelation. Pion, LondonGoogle Scholar
- Cliff AD, Ord JK (1981) Spatial processes: models and applications. Pion, LondonGoogle Scholar
- Fischer, MM, Getis, A (eds) (2010) Handbook of applied spatial analysis: software tools, methods and applications. Springer, BerlinGoogle Scholar
- Getis A, Ord JK (1992) The analysis of spatial association by distance statistics. Geogr Anal 24:189–206. [Reprinted in: Anselin L, Rey S (eds) Perspectives on spatial data analysis. Springer: Berlin]Google Scholar
- Getis A, Ord JK (2000) Seemingly independent tests: addressing the problem of multiple simultaneous and dependent tests. 39th Annual meeting of the Western Regional Science Association, Kauai, HawaiiGoogle Scholar
- Levene H (1960) Robust tests for equality of variances. In: Olkin I (ed) Contributions to probability and statistics. Stanford University Press, Stanford, pp 278–292Google Scholar