Was that in 2007? In 2003 LINEST I get a zero quadratic coefficient, but a
constant of 4.19E+08.

I did assume that the OP's regional settings use a comma as the decimal
separator.

Jerry

"Jon Peltier" wrote:

Using LINEST, the predicted zeroth order constant is exactly zero, which is
because of Excel 2003's insistence that small values are really related to
roundoff errors.

....

- Jon
-------
Jon Peltier, Microsoft Excel MVP
Tutorials and Custom Solutions
Peltier Technical Services, Inc. - http://PeltierTech.com