forum.alglib.net
http://forum.alglib.net/

Variance and Standard Deviation as Methods
http://forum.alglib.net/viewtopic.php?f=2&t=452
Page 1 of 1

Author:  dmprantz [ Fri Sep 30, 2011 2:41 pm ]
Post subject:  Variance and Standard Deviation as Methods

Looking at the descriptive statistics code, I see that both variance and standard deviation are calculated, but neither is encapsulated into its own method. Since these are values that can be useful besides just determining skewness and kurtosis, I would like to recomment/request that one or two public static methods be added to the descriptivestatistics class which encapsulate the code which appears to be currently between lines 78 and 97 of the C# source.

If it were me, I would create two methods, one for variance and one for standard deviation, and have one call the other. This would esepcially work since variance does not appear to be used except to derrive standard deviation. It appears that your preferred way to do that might be to create a single method with ref parameters for both values. I could use either one, but would like them to be there.

Please let me know if you will do this, and how long until the next release. I would be glad to personally make the change to both C# and C++, but I don't see repository access, and it looks to be a bit more complicated than that. Thank you,

dmp

Author:  Sergey.Bochkanov [ Sun Oct 02, 2011 7:52 pm ]
Post subject:  Re: Variance and Standard Deviation as Methods

I've added this issue to Mantis, see http://bugs.alglib.net/view.php?id=403 for the record.

This fix seems to be very easy and useful, and it will be implemented very soon (either by me or another programmer, Anatoly). However, next release is expected somewhere in the November. In the second half of November, I think, so it may take a while for this fix to reach the public. I prefer to make "large" releases, i.e. releases with at least several significant improvements - and right now we just don't have enough of improvements to make new release :)

Author:  dmprantz [ Sun Oct 02, 2011 8:40 pm ]
Post subject:  Re: Variance and Standard Deviation as Methods

Thank you. I will wait until November.

dmp

Page 1 of 1 All times are UTC
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/