lør, 07 03 2009 kl. 22:46 +0100, skrev Søren Hauberg: > lør, 07 03 2009 kl. 21:30 +0100, skrev Jaroslav Hajek: > > My vague impression is that R has the NA skipping optional. Can anyone > > confirm? > > It seems that be default R behaviour is to raise an error if NA's are > observed. You can then select to have observations with NA's deleted. > Two ways of deletion is supported. If, e.g, 3 dimensional vector is > observed where one of the values are NA, then > > 1) the entire observation is deleted, or > > 2) only the actual NA is deleted. > > At least that's the impression I get.
Okay, I pressed 'send' before reading what I wrote. Assume you have 4 observation from a 2D space: 1 2 3 NA 4 5 6 7 In the above-mentioned method 1, you would compute the covariance of 1 2 4 5 6 7 and in method 2, you would include the 2nd observation when computing variance of the first dimension, but exclude this observation when computing the variance of the second dimension and the covariance between the first and the second dimension. Søren ------------------------------------------------------------------------------ Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise -Strategies to boost innovation and cut costs with open source participation -Receive a $600 discount off the registration fee with the source code: SFAD http://p.sf.net/sfu/XcvMzF8H _______________________________________________ Octave-dev mailing list Octave-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/octave-dev
