Prior to Excel 2003, LINEST and the chart trendline made the same mistake in
concept regarding R^2 when the intercept is forced to zero. In Excel 2003,
the concept was corrected in LINEST, but not in the chart trendline. Example
data with certified results are available from NIST

http://www.itl.nist.gov/div898/strd/...ATA/NoInt1.dat
http://www.itl.nist.gov/div898/strd/...ATA/NoInt2.dat

Jerry

" wrote:

On Feb 21, 5:06 pm, Jerry W. Lewis wrote:

Note that in Excel 2003 the chart trendline R^2 is still incorrect when the
intercept is forced to zero.

Jerry

So the R^2 value generated for the trendline -- the one Excel displays
next to the trendline along with the equation -- is off? Is the one
that LINEST spits out any more accurate? Thanks again for the help.