From: <sch...@us...> - 2012-03-09 20:03:56
|
Revision: 9803 http://octave.svn.sourceforge.net/octave/?rev=9803&view=rev Author: schloegl Date: 2012-03-09 20:03:49 +0000 (Fri, 09 Mar 2012) Log Message: ----------- explain relationship to corrcoef in help text Modified Paths: -------------- trunk/octave-forge/extra/NaN/inst/cov.m Property Changed: ---------------- trunk/octave-forge/extra/NaN/inst/cov.m Modified: trunk/octave-forge/extra/NaN/inst/cov.m =================================================================== --- trunk/octave-forge/extra/NaN/inst/cov.m 2012-03-09 16:13:10 UTC (rev 9802) +++ trunk/octave-forge/extra/NaN/inst/cov.m 2012-03-09 20:03:49 UTC (rev 9803) @@ -4,6 +4,14 @@ % NaN's are skipped, NaN do not result in a NaN output. % The output gives NaN only if there are insufficient input data % The mean is removed from the data. +% +% Remark: for data contains missing values, the resulting +% matrix might not be positiv definite, and its elements have magnitudes +% larger than one. This ill-behavior is more likely for small sample +% sizes, but there is no garantee that the result "behaves well" for larger +% sample sizes. If you want the a "well behaved" result (i.e. positive +% definiteness and magnitude of elements not larger than 1), use CORRCOEF. +% However, COV is faster than CORRCOEF and might be good enough in some cases. % % C = COV(X [,Mode]); % calculates the (auto-)correlation matrix of X @@ -26,7 +34,7 @@ % http://mathworld.wolfram.com/Covariance.html % $Id$ -% Copyright (C) 2000-2003,2005,2009 by Alois Schloegl <alo...@gm...> +% Copyright (C) 2000-2003,2005,2009,2011,2012 by Alois Schloegl <alo...@is...> % This function is part of the NaN-toolbox % http://pub.ist.ac.at/~schloegl/matlab/NaN/ Property changes on: trunk/octave-forge/extra/NaN/inst/cov.m ___________________________________________________________________ Modified: svn:keywords - Id Revision + Id This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site. |