[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[OctDev] Function mean() fails with a complex matrix

From: Michael D. Godfrey
Subject: [OctDev] Function mean() fails with a complex matrix
Date: Sat, 07 Mar 2009 13:27:52 -0800
User-agent: Thunderbird (Macintosh/20081209)

Jaroslav Hajek highegg at gmail.com wrote:

 Note that even
R, a statistical language, does not skip the NAs by default.
I would like to recommend that for statistics functions, R should
be treated somewhat like Matlab, i.e. compatibility with R should
be given high priority.  This is based on the following:

1. R is VERY widely used in the statistics community.  It is
   preferred to Matlab for Statistics.  As an example,
   practically everyone in the Stanford Statistics Department uses
   R for much of their computational work.  Matlab is now little
   used.  (Just for information, Stanford is widely viewed as the
   world's leading Statistics Department and methods and techniques
   developed there are used throughout the Statistics community.)

2. R is free software and the R developers will welcome collaboration.

Of course, there will be cases (probably quite a few) where Matlab
and R disagree on the definition and implementation of statistics
functions.  I do not see a simple way of resolving this in general.
It will often be true that the R implementation will be preferred by
statisticians, but other users may prefer Matlab compatibility for
convenience (at the cost of more accurate or informative results).

In any case, this should be given thought by the Octave developers,
so I am also posting this on the developer's list.

Michael Godfrey
(Stanford University)

reply via email to

[Prev in Thread] Current Thread [Next in Thread]