Further: “Our results suggest that the principal cause of the invalid cluster inferences is spatial autocorrelation functions that do not follow the assumed Gaussian shape”.This has your basic Magic Walrus error written all over it.
Here is a quick rule. Subdivide and conquer. The scans themselves result from transformation theory to relate the radial and lateral density functions, and these will be divide and conquer algorithms. Whether done in F or XZY domain, they will be finite, matrix operations, yielding a finite probability distribution. The research has to stay finite, use sample data convolution, computer intensive. But then one has to watch out for Ito's conditions to generate valid distributions.
Why Bell Curve?
A distribution of events that deliver a Bell shaped, especially if centered, is a distribution from some process that minimizes interactions with neighbors. In economic terms this means most trucks are mostly full, supply and demand optimally in phase.
No comments:
Post a Comment