Environmental Engineering Reference
In-Depth Information
Fig. 6.7 Residual diagnosis graph. a IDW; b Topo to Raster; c Spline; d Circular; e Exponential;
f Gaussian; g Stable; h Spherical
Table 6.5
Skewness and Kurtosis values for all interpolation techniques
Variable
Skewness
Kurtosis
IDW
6.525
261.72
Topo
6.402
280.688
Spline
6.563
282.067
Circular
0.599
4.941
Exponential
-1.69
12.75
Gaussian
0.071
0.193
Stable
-4.702
85.183
Spherical
-6.367
38.645
skewed right means that the right tail is long relative to the left tail. The Standard
residual diagnosis graphs show that the best results in terms of normal distribution
belong to Gaussian method, because all scattered values spread close to zero
(Skewness is 0.071) between the ranges of -0.5 and 0.5 (Fig.
6.7
).
Kurtosis is a measure of whether the data are peaked or flat relative to a normal
distribution. The dataset for Circular and Exponential with high kurtosis of 4.941
and 12.75 tend to have a distinct peak near the mean, decline rather rapidly, and
have heavy tails. Datasets with low kurtosis belong to Gaussian with the value of
0.193 tend to have a flat top near the mean rather than a sharp peak.
Search WWH ::
Custom Search