Skip to main content


Figure 4 | International Journal of Health Geographics

Figure 4

From: Accounting for regional background and population size in the detection of spatial clusters and outliers using geostatistical filtering and spatial neutral models: the case of lung cancer in Long Island, New York

Figure 4

Different neutral models for male lung cancer. The fill color in each ZIP code represents the simulated SMR, with green indicating relatively low SMR and purple representing relatively high SMR (categories correspond to deciles of the histogram of simulated rates). Simulated maps (realizations) of the spatial distribution of lung cancer SMR data are generated under the assumption of complete spatial randomness (Model I), or created using geostatistical simulation in order to reproduce the spatial autocorrelation displayed by observed rates (Model II) as well as the regional background, i.e. SMR local means (Model III).

Back to article page