This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: bug in covariance_source.cpp??
- From: Brian Gough <bjg at network-theory dot co dot uk>
- To: Juan Jose Gomez Cadenas <gomez at ific dot uv dot es>
- Cc: <gsl-discuss at sources dot redhat dot com>
- Date: Sat, 12 Apr 2003 11:50:29 +0100
- Subject: Re: bug in covariance_source.cpp??
- References: <32836.147.156.135.181.1045853361.squirrel@ific.uv.es><32831.137.138.205.82.1049968448.squirrel@ific.uv.es>
Juan Jose Gomez Cadenas writes:
> In covariance_source.cpp, the function gsl_stats_covariance
> calls the function covariance_m but I believe it should
> call the function covariance
The _m function takes a precomputed mean, rather than a "known
mean" (in the statistical sense), so I believe the factor of
1/(n-1) is ok and should be the same for both cases. Maybe
this should be clarified in the documentation though.
--
Brian