S. Karthi,
I strongly recommend that you do NOT use the ATP for anything having to
do with statistical analysis, regardless of the version of Excel you
have. Your version, and all earlier ones, uses very poor algorithms
that, while mathematically OK on computers with infinite precision and
memory, stink on what we currently have.
Grab a good textbook on econometrics, say, such as that by Jerry Green,
to understand what's up with ANOVA; there are a lot of excellent
resources out there (and, unfortunately for me, in someone else's
garage--- I'm still moving). Do a Google search for the last 3 years
within the *.excel.* groups for stuff by Jerry W. Lewis; he has posted a
lot of excellent work-arounds for Excel's stat-shortcomings, and I'm
pretty sure he has posted one for ANOVA as well. In other words, with
Excel, you should definitely work it out from the core capabilities,
being aware that Excel's VAR-family of functions is flawed in your
version, and all earlier ones. Jerry has also posted nice work-arounds
for these problems.
Easiest route? Download R (an open-source version of S) which is free,
and will do the calcs right. If you are at all into stats, knowing R, or
S, or S-Plus (nice Windows-based version of S with many additional
features), or MatLab or even Mathematica will be a heck of a lot better
than the (currently) ridiculous stuff Excel does with stat-related
things (with the exception of its Chart Trendline, and a few recent
improvements in some of the worksheet functions with v. 11, but *not*
the ATP)
HTH
Dave Braden
MVP - Excel
