The correlation distance
[[delta].sub.m] is used in .
The average correlation distance
is about 7 nm and 4 nm in horizontal and vertical directions, respectively.
. The correlation distance
computed between two n-vectors u and v is defined as
3, which was calculated on the basis of a 2[degrees] x 3[degrees] grid system, 150% of the typical grid separation of 250 km translates to an e-folding correlation distance
of approximately 0.38 x [10.sup.3] km, for which choosing [[alpha].sub.FDR] = [alpha] produces only very slight test conservatism.
The range distance can be referred to as correlation distance
, over which lag the values of the variable are not correlated (independent).
We defined a linkage when two acupoints were closer than 0.1 correlation distance
(the threshold is defined by considering the distribution of correlation distances
between every acupoint; the lower 1 percentile of the distribution is below 0.1 correlation distance
A correlation distance
is defined as a dissimilarity metric derived from Pearson's correlation coefficient  given as
Multipath fading is neglected at high frequencies as it is averaged out due to much shorter correlation distance
as compared to shadow fading.
Thus, for an approximate estimate of the correlation coefficients [[rho].sub.ij] it is possible to define a constant correlation distance
d and to assume that [rho] linearly varies from 1 to 0 as the distance between the two samples goes from 0 to d + 1.
The tested similarity measures include Manhattan distance, Euclidean distance, L-infinity (Maximum) norm, Mahalanobis distance, cosine distance, standardized Euclidean (Seuclidean) distance, and correlation distance
. Let X and Y be two feature vectors with dimension n.
Moreover, the importance of the correlation distance
between two adjacent points of the spatial grid has been investigated.