The point about 6th order fits is not that Excel's differ from those
calculated by other packages, but that these fits are not based on any
models of physical behavior, but rather on conforming "nicely" to the
measured points. If you are interpolating by using the fit, you will get
reasonable predictions. If you extrapolate too far beyond the range of
observations, you could be very far from the true behavior.
- Jon
-------
Jon Peltier, Microsoft Excel MVP
Tutorials and Custom Solutions
Peltier Technical Services, Inc. -
http://PeltierTech.com
_______
"daz" wrote in message
...
Thanks for your reply Jon,
I was going to add that I'm already using the linest function for the the
number crunching. However, why mess up a good feature. In 2000 I could
copy
and paste the contants quickly into a cell to generate a comparison with
the
raw data. Will this be fixed or explained?
I have read a few comments about the six order poly should only be used
for
making things pretty. I have to disagree. If the number is set to
scientific
and 8 sig figs are used, a trendline can replicate raw data points very
accurately. I generate calibration curves for venturi nozzles from NEL
empirical data using excel 2000. The coefficients generated are identical
to
other software analysis tools I use for programming.
regards
Daz
"Jon Peltier" wrote:
I've heard of issues with Excel 2007's display of the polynomial formula.
You could get the values in the worksheet using LINEST, as Bernard
explains:
http://people.stfx.ca/bliengme/ExcelTips/Polynomial.htm
Keep in mind that a sixth order fit is generally not physically
meaningful,
but may only serve to make the line look nice. Depending on the data you
are
plotting, you may find it more sensible to use a relationship that
matches a
theoretical model.
- Jon
-------
Jon Peltier, Microsoft Excel MVP
Tutorials and Custom Solutions
Peltier Technical Services, Inc. - http://PeltierTech.com
_______
"daz" wrote in message
...
Using 2007: when openning up a 2000 spreadsheet, a chart that used to
display
the equation for a 6th order poly only displays the the last 3
constants.
I
have looked at the trendline formating tools available and every thing
is
set
correctly. The trendline is drawn correctly and changes, as you would
expect,
when I increase the poly order.