I think his point was that Microsoft's reputation for accurate statistical
calculation is quite poor.

Logigally one might assume that functionality provided by the processor
(trig functions, etc) was not re-invented by Microsoft or other software
vendors. As such, they should be quite good. Limited testing suggests that
they provide machine accuracy (or something very near to it). However
functions that MS had to write from scratch may have algorithm issues. As
the OP correctly notes, there are issues with the accuracy and working range
of most probability and statistics functions as well as Engineering
functions.

Excel 2003 improved the accuracy of most statistics functions (but
introduced a new bug into LINEST, where coefficients of exactly zero are not
to be trusted). Excel 2003 improves the accuracy and working range of
NORMDIST, NORMSDIST, NORMINV, and NORMSINV. 2003 improves the numerical
inversion of other continuous distributions, but that is of limited value
since the accuracy of their distribution functions has not been improved.
2003 claims to imrpove the working range of discrete probability
distributions, but does so at the expense of accuracy, for example there are
POISSON probabilities that 2003 calculates that are wrong without warning by
as much as 0.5 (huge since 0<= probability <=1). Accuracy and working range
issues with engineering functions have yet to be addressed at all by MS.

Excel 2007 fixes the bug in LINEST that 2003 introduced, but otherwise has
made no improvements in accuracy that I was able to find.

Jerry

"JLGWhiz" wrote:

Personally, I rely on the reputation of Microsoft.

"ForIVD" wrote:


After reading all the research on the accuracy of statistical
functions in Excel how can the end-user be certain the non-statistical
functions are correct? Is anyone aware of a workbook that compares
output of Excel functions to known results to verify accuracy? Does
anyone know the name of a company that has done this kind of work?

Thanks,
David