Skip to main content

Table 4 Performance comparison of alternative kriging estimators: mean square standardized residual and goodness of uncertainty models.

From: Geostatistical analysis of disease data: accounting for spatial support and population density in the isopleth mapping of cancer mortality risk using area-to-point Poisson kriging

Estimators

Lung cancer

Cervix cancer

MSSR

Average

% best result

Average

% best result

Point kriging of raw rates

1.707

79

2.176

41

Point kriging of global EBS

5.941

0

4.566

0

Point kriging of local EBS

6.310

0

6.599

0

ATP Poisson kriging

1.827

21

1.776

59

ATP Poisson kriging (true γR(h))

1.081

 

1.086

 

GOODNESS STATISTIC

Point kriging of raw rates

0.918

78

0.868

27

Point kriging of global EBS

0.602

0

0.731

3

Point kriging of local EBS

0.580

0

0.616

0

ATP Poisson kriging

0.862

22

0.916

70

ATP Poisson kriging (true γR(h))

0.981

 

0.959

 
  1. Results obtained on average over 100 realizations generated for Regions 1 and 2. Poisson kriging was conducted with the semivariogram model derived through deconvolution or inferred directly from the simulated grid risk values (true point-support model γR(h)). Bold numbers refer to best performances (i.e. criteria closer to 1) outside the ideal case where the true semivariogram of risk is known. The second column gives the percentage of realizations where the particular method (except ATP kriging with true γR(h)) yields the best results.
Back to article page

International Journal of Health Geographics

ISSN: 1476-072X