Open AccessComputing the correlation and other things directly from the raw pairs
Author(s) -
David Brown
Publication year - 2007
Language(s) - English
Resource type - Reports
DOI - 10.2172/924950
Subject(s) - covariance , cartesian coordinate system , cartesian product , harmonics , correlation , computer science , raw data , mathematics , combinatorics , algorithm , physics , statistics , geometry , quantum mechanics , voltage
We want a faster and more robust way to compute the correlation, expanded in Spherical (or Cartesian) Harmonics. We also want to include the cross-{ell},m data covariance that are there, but currently ignored. We don't want to get bogged down in fancy binning in x-y-z or r-{theta}-{phi}, just r. Want to just look at C{sub {ell}m} to decide how many terms to keep--or better yet the pair distributions directly