OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office-comment message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Re: [office-comment] RSQ - Unstable Algorithm


I noticed that the algorithms used for the various OOo Calc functions 
have been recently improved/corrected [including RSQ() ]. While I 
sneaked through the code, it struck me the simplicity of the RSQ() 

Indeed, RSQ() is the *square of the Pearson coefficient*, so I wonder:
*Why does the OASIS document contain that awful formula  instead of the 
straightforward one*?

I verified in the last hour that the 2 formulas are indeed 
*mathematically identical*, so I am really puzzled. The square of the 
Pearson coefficient is simply:
r^2 = { sum[(Xi - Xm)(Yi - Ym)] }^2  / {sum[(Xi - Xm)^2] * sum[(Yi - 
Ym)^2] }

Indeed, just take the square of the Pearson coefficient!!!

[IF somebody is interested in the mathematical calculations, give me a 
short notice, BUT I'll need a couple of days to type it on the computer.]

[The used formula did *MISLEAD* me, too. I thought it is some really 
weird function.]

Is there any reason to use that complex and incomprehensible formula, 
instead of the simple and straightforward one?



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]